Skip to main content

Full text of "Gregorian chant, a textbook for seminaries, novitiates and secondary schools"

See other formats

^ » «»»■» * » » » 3 » 


handle this volume 

with care. 

The University of Connecticut 
Libraries, Storrs 









Digitized by the Internet Archive 
in 2013 







Teacher of Music 
Cathedral College, Brooklyn, N. Y. 

Published by 




Bishop of Brooklyn * i ~* 

Nihil Obstat 


Censor Ltbrorum 


JANUARY 27, 1945 

Desclee and Company of Tournai, Belgium, has granted permission to the 
author to use the rhythmic marks in this textbook. 

Copyright, 1945, by Gregorian Institute 
printed in u.s.a. all rights reserved 

Dedicated to 

Bishop of Brooklyn 


In the following pages Father Klarmann presents a clear, orderly, 
systematic treatment of liturgical chant. 

At the very beginning of his treatise he provides an explanation 
of certain fundamental terms, such as notation, signs, rhythm, 
chant structure, etc., which is very serviceable in preparing the 
reader for the fuller development of the general theme in the sub- 
sequent chapters of this book. 

With the same thought and purpose the author more particu- 
larly gives an early definition of the chief subject of discussion, 
namely, chant, which he defines, in the usually accepted sense, as 
liturgical music in the form of plain song, which is monophonic, 
unaccompanied and free in rhythm. Very interestingly also chant 
structure is explained. The author then proceeds to record the his- 
torical development of chant at least in its salient features. 

It is readily understood of course that the Infant Church could 
not promote a notable advancement in liturgical music during the 
period of ruthless persecution. And still it seems quite certain that 
even in the catacombs hymns were used in connection with religious 
worship. Tertullian and later Eusebius, Augustine and Jerome more- 
over give testimony regarding the traditional custom of the Church 
in sponsoring antiphonal and responsorial chant as well as the sing- 
ing of hymns in relation to sacred service. 

After the period of early trial and tribulation the Church was 
free to foster a definite type of liturgical chant which in content, 
form and spirit would be appropriate particularly in praising and 
glorifying God. And the Church, as history convincingly testifies, 
took full advantage of this opportunity. 

A word of special tribute for this progress in the development of 
chant is due to St. Ambrose who contributed so substantially to 
the realization of the aim of the Church in this regard by systema- 
tizing and establishing within his own Diocese of Milan what is 
known as the Ambrosian Chant. Later Pope Celestine introduced 
Ambrosian Chant to Rome. 

We know, moreover, that during the fourth century the Papal 
Choir was established as well as the Schola Lectorum in which boys 
destined for the Papal Choir were taught the chanting of the les- 
sons of the Mass. 

Notwithstanding the prominence and importance which chant 
thus attained in liturgical service it gradually began to acquire very 
objectionable features. In his book entitled: "Church Music — Its 
Origin and Different Forms " Janssens refers to this unsatisfactory 


change in these words: ''People did not relish the chant, and music, 
substituted to please them, degenerated into worldly, light and 
frivolous songs which evoked the bulls of popes and the decrees 
of councils." 

To Pope Gregory the Great a fairly reliable and constant tra- 
dition, as the author of this book indicates, ascribes the necessary 
and authoritative standardization of the Roman Chant. Pope Pius 
XI made impressive reference to this fact when he wrote: "It was 
in the Lateran Palace that Gregory the Great, having made his 
famous collection of the traditional treasures of plain song, editing 
them with additions of his own, wisely founded his great Schola in 
order to perpetuate a true interpretation of the liturgical chant." 
(Divini Cultus sanctitatem. Dec. 20, 1928.) 

Father Klarmann also alludes to the fact that during succeeding 
centuries the norms, earlier established in, this matter, were either 
forgotten or disregarded and musicians again changed the chant to 
satisfy popular taste and to provide more entertaining melodies. 

The Council of Trent consequently expressed a strong protest 
against the custom of using secular themes in the composition of 
sacred music and also against the practice of sacrificing the test for 
melodic effect. 

Later still corrective measures were necessary and in this regard 
we may note that Father Klarmann gives deserved recognition to 
the Benedictine monks of Solesmes, who, towards the end of the 
nineteenth century, started a movement to restore the simplicity, 
purity and reverence of the chant. 

It is perhaps useful to recall that Pope Pius X declared the 
Solesmes version of the chant to be the official version and it is ap- 
propriate also I am sure to point out that schools of liturgical music 
immediately sprung up in various dioceses to promote systematically 
the reform in church music thus decreed. 

Pope Pius XI stressed the requirements of the Motu Proprio 
of Pius X and exhorted the clergy and laity to restore approved 
liturgical music to churches. 

Now the question may arise why does the Church establish her 
own form of music and insist upon its conformity to certain norms? 
The reason undoubtedly is that the chant is associated with, re- 
ligious functions and more particularly with the Mass, the center of 
Catholic worship, and therefore the chant must be in conformity 
with the place, time and purpose of Divine worship. It should be 
therefore sacred and not profane music. 

Then again true religious music is but an exalted prayer — an 
exultant expression of religious feeling. In making this observation 
there is suggested at once an appropriate quotation from St. Paul: 
"Speaking to yourselves in psalms and hymns and spiritual canticles, 
singing and making melody in your hearts to the Lord." (Eph. 
V: 19.) 


And in further reference to the prayerful function of liturgical 
chant Sister Mary Gael, of Marygrove College, very pertinently 
and piously observes: "It is not difficult to understand how the 
simple flow of pure melody carrying words of the sacred text can 
speak to the soul of God and then lead the soul to speak to God in 

Sometimes perhaps the impression might be developed that 
the church chant represents a type of inferior music at least in com- 
parison with secular musical compositions. 

In dealing with such an unfortunate and incorrect impression 
it might be well to quote Professor Willi Apel, of Harvard Univer- 
sity, who, in his Dictionary of Music, published in 1944, remarks: 
"Whereas formerly musicians looked disdainfully on Gregorian 
chant, particularly because it 'lacks' harmony, it is now becoming 
more and more fully recognized as an unsurpassed treasure of 
purely n Iodic music. In particular, its freely flowing rhythm, far 
from being chaotic, shows subtleties of structure and organization 
which are doubtless superior to the comparatively platitudinous 
devices of rhythm in harmonized music, with its meter, measures, 
beats, regular phrases, etc." 

Then again competent musicians have admitted the superior 
quality and in fact the supreme excellence of church chant. I refer 
to such capable critics as Witt. Gevaert, Halevy, Mozart, Berlioz. 
{Catholic Encyclopedia.) Halevy, for instance, observes: "The chant 
is the most beautiful religious melody that exists on earth." Then 
there is Mozart's statement, that he would gladly exchange all his 
music for the fame of having composed the Gregorian Preface. 

In view therefore of its use in religious worship; of its formal 
approval by the Church and of its inherent simplicity, purity and 
excellence we may readily agree that Father Klarmann has ren- 
dered a highly meritorious service in trying to promote a truer 
understanding, deeper appreciation and more extensive use of 
Gregorian Chant. 

Bishop of Brooklyn 

Brooklyn, New York 
June 25, 1945 


If we were to adopt the medieval style of designating this book, 
we might entitle it "A Practical Commentary on all the Features of 
Gregorian Chant, with Special Attention Given to the Principles 
of Rhythm as Proposed and Formulated by the Monks of Solesmes." 
Using this modern method of introduction, however, we state that 
this is a textbook containing all the features of the chant, explained 
in such a way that a person with little or no musical training could 
acquire sufficient knowledge of the subject to be able to sing the 
chant with correctness, propriety and beauty. This book is intended, 
therefore, for those in seminaries, novitiates, secondary schools and 
other institutions of learning whose curriculum includes a study of 
the chant, as well as for choirmasters, their singers and for all others 
who may be interested in this branch of musical art. 

A cursory examination of the Table of Contents or of the book 
itself will reveal that it is divided into three parts. The first three 
chapters of Part I are a technical explanation of the signs and 
figures of the chant as they appear in print, and of the means by 
which long notes are designated. The following nine chapters deal 
with the theory of rhythm, its analysis, its purpose and the method 
by which this purpose is achieved. Illustrations are taken, for the 
most part, from the Requiem Mass and the Ordinary. Chapters IV 
to XII follow in logical order and it is suggested that they be studied 
or read to best advantage in the sequence in which they appear. 
The remaining two chapters of Part I present incidental, but none 
the less important, reflections on the chant. Part II includes a 
treatment of the more difficult or more advanced aspects of chant 
(modes, psalmody), of Latin pronunciation, indispensable in the 
proper singing of the chant. Part III includes treatises on the 
relation between the Mass and the chant, the history of church 
music, and finally the rules of the Church regarding music during a 
liturgical function. 

Throughout the textbook we have tried as far as possible to 
avoid the use of technical language. For example, the term "fall 
group" is used instead of "thesis" and "rise group" instead of "arsis." 
Even in treating the nature of a Latin word the term "rise" is some- 
times adopted for the accented syllable and "fall" for the grave 
syllable of a word. The translations of the Greek and Latin names 
for the neumes are supplied as a teaching aid since the picture con- 
jured up by the English word is more graphic than the foreign term, 
especially in the teaching of younger students. 


The author wishes to express his gratitude to the monks of 
Solesmes, particularly to Dom Gajard, who commended upon the 
ideas of rhythm contained in this textbook which was written in 
substance during the author's stay at the monastery; to Right 
Reverend Monsignor Lawrence Bracken, Chairman of the Diocesan 
Commission for Church Music in the Diocese of Brooklyn, for his 
encouragement and valuable advice; and, lastly, to Doctor Clifford 
Bennett, without whose cooperation this book would never have 
been as practical as it may be and without whose interest it might 
never have been published. 

A. F. K. 



Foreword iii 

Introduction vii 

Part I 

Chapter Page 

I. . Gregorian Notation and Signs 1 

II. Neumes or Combinations of Notes 11 

III. The Duration of Notes 18 

IV. The Nature of Rhythm (23) 

V. The Development of Rhythm (27 

VI. The Purpose of Rhythm 34 

VII. The Uniting of Notes in Neumatic Chants 36 

VIII. The Uniting of Notes in Syllabic Chants 42 

IX. The Uniting of Groups of Notes 54 

X. The Uniting of Groups into Divisions and of Divisions 
into Phrases 58 

XI. The Joining of Divisions 65 

XII. The Application of the Principles of Rhythm to Special 

Figures . . . * 67 

XIII. Directing the Chant 75 

XIV. The Application of Gregorian Principles to Modern 
Music 77 

Part II 

XV. Gregorian Modes 82 

XVI. Psalmody 90 

XVII. Latin Pronunciation 104 

Part III 

XVIII. The Mass and the Chant 113 

XIX. The History of Church Music 121 

XX. Church Music Legislation 131 

General Index 144 


f 1-4 




1. Gregorian melodies are written on a staff of four lines and 
three spaces, while modern music is written on a staff of five lines 
and four spaces. The notes are for the most part square in shape, 
preserving their original character of a dot or period made with a 
quill. Notes are found also in other forms for different purposes and 
effects, as will be explained in subsequent chapters. 

2. The notes indicated on the lines or in the spaces of the staff 
represent those of the spl-fa system. These notes are conventionally 
called: do, re, mi, fa, sol, la and ti. We shall often have occasion to 
replace them by numbers in such a way that 1 will represent do; 
2, re; 3, mi; etc. Notes above the middle octave will be distinguished 
by a dot over them: 6 7 12; those below this octave, by a dot 
under them: 6 7 12. 

3. The sol-fa scale consists of eight notes comprising an octave. 
Each of the eight notes is one whole tone from the preceding, ex- 
cept fa and do which are only a half tone higher than mi and ti 
respectively. The following scheme shows the scale steps and the 
number designation to«be used throughout this book: 

-j tone 

— r- tone 



mi fa 



ti do 



. 3 4 



7 i 

4. The notes of the sol-fa scale are relative in pitch; that is, they 
are gauged according to the pitch chosen for the key-note or the 
tonic. They do not represent definite notes on the piano as do the 
notes on the modern staff. This means that the note re, for example, 
does not refer to a definite note on the piano, but may be any note 
depending upon the note chosen for do. Hence, if the piano note F 
were chosen for do, then G would be re, A would be mi and B[? 
would be fa, etc. If, however, E were chosen for do, then F# would 
be re; G# would be mi, and A would be fa. The reason for the use 
of sharps lies in the fact that the notes E and F on the piano are 
only a half tone apart, whereas in the sol-fa scale do and re are a 
whole tone apart. In order, therefore, to play the whole tones do 
and re when E is chosen for do, F# must be used for re. From this 
we may readily deduce that any modern key may be used to play 


a Gregorian melody provided the intervals on the piano are made 
to correspond exactly to the intervals of the sol-fa scale. The key in 
modern music is determined according to the note chosen for do. 
In speaking of the key of F, we simply mean that the F is do in 
that key. For a more detailed exposition of the scales and tonalities 
of the chant and appropriate keys for the melodies, see Chapter XV. 

5. The chant notation uses one of two clefs to designate either 
do or fa on the staff. From these notes the other notes may easily 
be deduced. The two clefs are known as the do clef, indicating the 
position of do, and the fa clef, indicating the position of fa: 


do clef fa clef 

Fig. 1 

6. The do clef may appear on the second, third or fourth line 
(from the bottom), but is most frequently found on the third or 
fourth line; the fa clef may appear on the third or fourth line, but 
it is most frequently found on the third. Neither clef ever appears 
in any space or on the first line from the bottom. The following 
staffs will show the various positions of the clefs and their influence 
on the notes: 



























fa sol la ti do re mi fa 
4 5 6 7 i 2 3 4 

Fig. 2 

















/a it ^o r£ mi /a jo/ /a 
6 7 12 3 4 5 6 

Fig. 2 (cont.) 

7. The choice and position of the clef are determined by the 
range of the melody. That particular clef is chosen and is so located 
that the melody will fit conveniently within the lines of the staff. 
When a melody exceeds the limits of the staff, leger lines are used 
as in modern music. 

8. There are no sharps in the chant and only one note may be 
flatted. That note is ti; when flatted it is called te. When te is repre- 
sented in number, the 7 is crossed (X) to distinguish it from ti. 

9. Te forms a half tone between 6 and t and a whole tone be- 
tween X and 1. The succession of 6* 1 has the same relation as 34 5, 
whereas the succession 5 6t corresponds to 2 34. Since ti is the 
only note that can be altered, composers often resorted to trans- 
position when they wanted a whole tone between 3 and 4. The 
transposition could be effected either by changing the position of 
the do clef, or by using the do clef instead of the fa clef. In the 
following illustration at (a), the desired whole tone between 3 4 is 
effected by a change of the position of the do clef and the flatting 
of ti. At (b), however, the do clef is used instead of the fa clef, and 
the ti is flatted. 

(a) (b) 


(a) : 



Fig. 3 


1 10-11 






Fig. 3 (cont.) 

Very often, however, the use of the flat throughout the entire 
melody is indicative of a transposed melody (see par. 178). 

10. The change of ti to te is indicated on the staff by the flat 
sign i\) and effects only the ti over that word on which it occurs. 
Te automatically becomes natural if the flat is not repeated over 
the next word. 'See Fig. 4 [a]). If the word extends over a divisional 
sign (par. 12) the flat likewise loses its force beyond that sign un- 
less it is repeated (b) : 





super tcr-ram. 

(b) 3 





♦" r ! *V. « 

Tu glo- 

n- a 

Fig. 4 

If te is to be naturalized before the end of the word or before the 
divisional sign, the common natural sign (s) is used as indication. 
Ti remains natural, then, until the flat sign is definitely repeated. 

11. The notes are commonly square in shape, having originated 
in all probability as dots made with a quill. There are. however, 
different shapes of notes to indicate different purposes and effects: 

(a) the square punctilio ■ 

(b) the inclined punctum ♦ 

(cj the virga or stem ^ 

Fig. 5 

(d i the quilisma 1 

(e) the liquescent ^ 

(f ) the guide J 


(a) The square note is the ordinary and most common of all the 
notes. It is used either alone as a punctum or in some combina- 
tions (neumes) as will be shown in the next chapter. 

(b) The inclined note is used only in descending passages in combi- 
nations. It does not differ in time value from the square note. 
The reason for its inclined shape lies in the fact that the ancient 
scribes slanted the quill when writing such descending notes. 

(c) The virga or stemmed note is a note with a vertical line at its 
right or left. This note has no greater time value than the 
square note. The stem indicates a comparatively high note and 
is a remnant of the accent sign (/) employed today to indicate 
the acute accent of a word. Formerly every syllable of the 
word was marked with either the so-called 'acute' (/) or 'grave' 
(\) accent: 

Fig. 6 

Later, however, when the grave accent was replaced by a dot 
made with a quill, the acute accent was replaced by the stem- 
med note which is the present virga: 


Fig. 7 

Hence the virga is nothing more than a relatively high note, 
even as the accented syllables are the high syllables of the word. 

(d) The quilisma and 

(e) The liquescent are never found as single notes apart from the 
combination of notes; their purpose is explained in the next 
chapter (pars. 35, 43). 

(f) The guide or custos, is a small note of great assistance to the 
singer. Its various functions as a guide note are illustrated in 
the following example. At (a) it indicates the first note of the 
next line; at (b) it indicates the pitch of the next note follow- 
ing a division mark where the clef has been changed. 

C y 1 c — i 

(a) J I (b) ftfrfr 

1 ■ ■ ■ 

Propter ^. C6rpo-ra 

Fig. 8 


12. The melodies of the chant are divided and subdivided into 
musical sections similar to the clauses and phrases that make up a 
sentence. These divisions are determined by the nature of the 
melody and text. The marks indicating them may rightly be called 
'musical punctuation marks.' Each division is an entity in itself 
although it may be part of a larger unit, just as a phrase or clause 
forms part of a sentence. These divisions of the chant are designated 
by vertical lines cutting one or more lines of the staff: 

(a) (b) (c) (d) 

double phrase phrase member incise 

mark mark mark mark 

Fig. 9 

(a) The double phrase mark is a double line cutting all the lines of 
the staff; it represents the largest and most important division 
of a melody. It marks the end of a melody and the total comple- 
tion of the text. When it occurs within a melody, however, (as 
between the sections of a Kyrie or a Gloria or Credo) it denotes 
the place where the choirs are to alternate. In the Gradual 
chants it marks the place where the solo singer is to take up the 
melody, a place also designated in the text by the versicle mark 
(See Fig. 8 [b]). 

(b) The phrase mark is a single vertical line cutting all the lines of 
the staff. It represents the melodic and often textual completion 
of a section or division. It may be compared with a period in 
grammar in that it marks the end of a musical sentence and im- 
plies a completion that requires a definite pause. It demands a 
definite softness and slowness on at least the two last groups, and 
a complete silence of two beats before the next phrase is sung. 

(c) The member is so called because it is a member or part of a 
phrase. It implies no completion of melody or text but functions 
as a convenient punctuation of the phrase. It calls for a slight 
retard but implies no interruption. If breath must be renewed 
at this place, it should be 'stolen', as it were, from the previous 
beat in order to avoid a break in the singing. 

(d) The incise is the smallest musical subdivision. It subdivides the 
member just as the member subdivides the phrase. Since the 
subdivision is subordinate it needs only the slightest retard and 
definitely no interruption. It may be compared to the comma 
in writing. 

13. Occasionally a comma appears on the fourth line of the 
staff. Practically speaking, it differs little from the foregoing incise 
mark and is subject to personal interpretation. 


1f 14-15 

14. In general, melodies are not divided into equal parts as 
though one incise or member were always as long as the other. The 
divisions are almost never equal or regular throughout — the Requiem 
Offertory is a good example of irregular division. The first phrase 
is made up of three members, each of which embraces two incises; 
yet the second phrase is formed of three members which are un- 
divided; the third phrase is composed of two members, the second 
of which is subdivided into two incises, and so forth. The following 
diagram illustrates these subdivisions: 

1st member 

Requiem . 
Offertory x 

1st phrase • 

2nd phrase 

3rd phrase 

1st incise : Domine Jesu Christi, 
2nd incise : Rex gloriae, 
1st incise: libera animas 

2nd member^ 2nd incise : omnium fidelium de- 

1st incise: de poenis inferni, 
2nd incise: et de prof undo lacu: 
libera eas de ore leonis, 
ne absorbeat eas tartarus, 

3rd member: ne cadant in obscurum: 

1st member: sed signifer sanctus Michael 
. fist incise: repraesentet eas 

2nd member U nc i incise: in lucem sanctam. 

3rd member 

1st member: 
2nd member 

In the hymn Adoro Te, .however, we have an example of symmetrical 
division. Each stanza is made up of two phrases, composed of two 
members, each of which is composed of two incises. Each division 
is equal in length to its corresponding member. Each stanza is 
divided as follows: 


1st phrase 

2nd phrase 

1st member 
2nd member 
3rd member 
4th member 

fist incise: 

\ 2nd incise: 

3rd incise: 

4th incise: 

(5th incise: 

\6th incise: 

7th incise: 

8th incise: 

Adoro Te, devote , 
latens Deitas, 
'Quae sub his figuris 
vere latitas: 
Tibi se cor meum 
totum subjicit, 
quia Te contemplans 
totum deficit. 

15. The ictus in print is a short vertical line over or under a 


Fig. 10 

Its rhythmical significance is explained in Chapter V (see par. 72). 
Let us say, for the present, that the note on which this sign appears 
is the rhythmic bearer of the chant. In theory, it is the first note of 
a group; in practice, it represents a fall note of the rhythm. Not all 
the fall notes are marked since the majority are self-evident from 


the rules laid down in succeeding chapters YII and YIII . Those, 
however, that carry the sign are so designated to clarify any doubt, 
to simplify the reading, or to effect a definite rhythmic nuance. It is 
included here merely as one of the signs used in chant books. 

16. The episema is a horizontal line over or under one or more 
notes and indicates that the note or notes are to be prolonged be- 
yond their usual length 'see par. 145 . 



— V 







• V 

Fig. 11 

17. The asterisk *) suggests several directions: 

a It indicates the end of the intonation when it appears at the 
beginning of a selection. In the following illustration, the chanter 
intones the melody up to the asterisk, at which point the choir 

Intr. £ 


equi- em * ae-ter- nam 

Fig. 12 

(b) When it occurs toward the end of a selection, it denotes the place 
where the choir is to re-enter: 

di- ne * per-fru- i. 

Fig. 13 

(c) When it appears alone, dividing the last Kyrie into two parts, 
it indicates that the section of the choir whose turn it is to sing 
that particular Kyrie. sings the whole of it. and that the other 
section is to enter at the asterisk: 

* ■ ■ Si 

i ■ *V\ 

■ ■• ♦* 

♦• ■ ■ ■ ■' 

■ ■ 

■ • 

Ky-ri- e * e- ie- i-son. 

Fig. 14 


But if the Kyrie is divided into three parts by a single and a 
double asterisk, the first section of the choir sings to the first 
asterisk, the other section sings to the second (from the single 
to the double asterisk) and all sing from the double asterisk to 
the end: 

e * gT3s iiffk 

'♦♦• '** T *1% 



Ky- ri- e * ** e- le- i-son. 

Fig. 15 

(d) When the asterisk occurs in the middle of a psalm verse, it does 
not call for a change of choir, but indicates the place for the 
mediant cadence (see par. 198.). 

S ... la . 

■ ■ 

vo-tum in Je-ru- sa-lem: * exaudi o-ra-ti- 6-nem 

Fig. 16 

18. The letters ij or iij, found after the Kyrie s, Christes or the 
Alleluia's, are signs of repetition. The ij indicates that what pre- 
cedes is to be sung twice while the iij calls for a three-fold repetition. 
The letters are probably an abbreviation of the Latin word idem 
meaning 'the same.' Hence Kyrie eleison. iij. indicates that Kyrie 
eleison should be sung three times. 

19. Chant melodies are classified according to the richness of 
melody on the syllables of the text : 

(a) Syllabic chants are those in which, for the most part, there is 
one note to the syllable (The Psalms, Credo chants, sequences, 
some hymns, for example, the Adoro Te, the Preface and the 
Pater Noster are syllabic in character) . 

(b) Neumatic chants are those in which most of the syllables are 
adorned with neumes or combinations of notes (some Gloria 
chants, antiphons, arid hymns such as the Verbum supermini, 
are neumatic in character). 

(c) Melismatic chants are those in which most syllables are adorned 
with many neumes. This is the richest kind of chant (the 
Graduals, Tracts, and Alleluia verses are examples of melis- 
matic chant). 

20. Every Gregorian melody, in fact every piece of vocal music, 
is composed of three elements: melody, rhythm and text. Each of 
these elements has its place and is governed by separate laws: the 


rights of each are for the most part observed. Circumstances arise, 
nevertheless, which necessitate the precedence of one over the other 
two. For example, the natural treatment of a word may demand 
that the last syllable be lower in pitch than the accented syllable. 
Sometimes, however, the melody and word are so constituted that 
the last syllable will occur on a high melodic note. In this case the 
melody asserts its right over and above the word or text. So, also, 
can rhythm assert its right over the melody or text. A rhythmic 
rise should be higher in melody than its rhythmic fall. But we often 
find that the note on which the rhythmic rise occurs is lower in 
melody than its rhythmic fall. Then again, the last syllable of a 
word should be set to a rhythmic fall note since the last syllable is 
the end of the word and the fall note, the end of the rhythm. But 
still, the rhythm need not respect the word in all cases; it may, 
therefore, set a rhythmic rise on the last syllable of a word. It is 
our duty to heed the natural demands of each in so far as it is pos- 
sible to do so. The recognition of such demands will be taught in 
later lessons. 

21. The melody is designated by the notes on the staff and indi- 
cates (1) the pitch of the notes and (2) the number of notes to be 
sung on each syllable. The pitch is indicated by the position of the 
notes on the staff. The notes are so printed over each syllable that 
there is no difficulty in determining the number and kind to be sung 
on the syllable. 

22. The reading of the text should present no difficulty. The 
language used is Latin, though Greek and Hebrew words also occur. 
The words Kyrie eleison and those heard on Good Friday {Agios o 
Theos, etc.) are Greek. Words like Hosanna, Alleluia and Amen are 
Hebrew in origin. Whatever their origin, however, all words are 
pronounced or sung according to the value of their letters in Latin. 
This is fully treated in Chapter XVII. The melody will, in the 
majority of cases, follow the natural rhythmic curve of the word, 
rising to the accent and falling to rest from it. But, as has been said, 
it will at times disregard the natural rights of the word and follow 
its own course. 

23. The rhythm is the principle by which the melody and words 
are given flowing motion. This motion is described by the figure or 
neume in which the notes are represented, by the occasional ictus 
mark (par. 15), by the various signs of length (par. 48) and by the 
divisional marks (par. 12). But these do not adequately indicate 
clearly all the important features of rhythm. Rhythm is perhaps 
the most important aspect of the chant, and undeniably it is a 
salient feature of any musical composition. Yet it is most difficult 
to consign to print. Furthermore, it is difficult to establish rigid 
laws for proper rhythmic interpretation, for it is the primary duty 
of rhythm to respect as far as possible the natural demands of 
melody and of text. In rendering the chant, therefore, we must be 
guided by the rule that rhythm, melody and text must be set in as 
close agreement as is possible. This principle is so important that 
the major part of this book is devoted to its explanation. 


1 24-27 



24. Whenever two or more notes are to be sung on one syllable, 
the notes are arranged above them in figures called neumes. The 
word neume is derived from the Greek, either from pneuma meaning 
'breath,' or from neuma meaning 'sign.' These figures or neumes 
illustrate the rhythm graphically so that after a little practice the 
rhythm can easily be read from the scheme in which the notes are 
arranged. In this respect the ancient Gregorian notation is far 
superior to our modern notation which has no neumatic figures. 
Amateurs of the chant should not be discouraged by the strange 
appearance of notes and neumes. A little effort expended in becom- 
ing acquainted with the ancient system will soon be rewarded with 
facility in reading the rhythm from the figures. 

25. Too much stress should not be laid upon the memorization 
of the Latin or Greek names of these figures. After all, the ability 
to sing them correctly is our aim. For those who prefer more simple 
terms the English derivatives of the original foreign-language names 
have been added. The use of the English terms is especially recom- 
mended for younger students of the chant, for it seems unreasonable 
to burden them with the task of learning foreign names that are 
unintelligible to them. The foreign names are retained, however, in 
order to satisfy the more facetious critics. 

26. A knowledge of the physical structure of the neumes is 
necessary for the complete understanding of Gregorian rhythm. A 
knowledge of the neumes is interrelated with a full knowledge of 
rhythm: it is difficult to explain one without the knowledge of 
the other. 

27. The neumes probably had their origin in the two gram- 
matical accent signs: the grave accent (>) and the acute {?). Both 
were made from left to right; the former described by a downward 
movement; the latter an upward movement. The acute accent was 
used for a higher note; the grave for a lower note. The combination, 
therefore, of a high note followed by a lower note was represented 
in the ancient manuscripts by the circumflex (A) ; whereas a low 
note followed by a higher was represented by its inverse (v)- When 
several notes descended from the higher or first note, the figure 
took the shape of A etc. Such, then, is the probable origin of the 
forerunners of our present neumatic system. 


If 28-33 


28. Neumes may be divided into three classes: (a) simple; (b) 
compound; (c) special. A simple neume is made up of either two or 
three notes; a compound neume is a simple neume that has been 
enlarged either by the addition of more notes, or by the combining 
of two or more neumes; a special neume is a simple neume that 
serves a special purpose. 

29. There are three simple two-note neumes: the podatus, in 
which the lower note is sung first; the clivis, in which the higher note 
is sung first; and the bivirga or distropha, in which the two notes 
are in unison. These simple neumes are illustrated below. 

30. The podatus or pes (from the Greek and Latin words for 
'foot') : a neume of two notes, of which the lower is sung first. 


2 3 

3 5 

2 6 

Fig. 17 

31. The clivis (from the Latin for 'incline') : a neume of two notes, 
of which the higher is sung first. This higher note always has a 
virga (par. 11 [c]), a trace of the original acute accent (see par. 11 
and 27). 

1 7 

1 6 

1 4 

Fig. 18 

32. The bivirga or distropha (Latin and Greek for 'double stem' 
and 'double comma' respectively) : the two notes of this neume are 
in unison or on the same pitch. The difference between the two forms 
lies in the fact that the bivirga is used when the two notes form an 
important rhythmic climax (see par. 124). 



4 4 4 4 

Fig. 19 

33. There are five kinds of simple three-note neumes: (1) in which 
all three ascend; (2) in which all three descend; (3) in which the 
middle note is highest; (4) in which the middle note is lowest; (5) in 


HEUMES 1134-35 

which all three are in unison. The first of these (three notes ascend- 
ing) subdivides into three classes: (a) in which all the notes are of 
equal length ; (b) in which the first note is prolonged ; (c) in which the 
second note is prolonged. Illustrations and further commentary on 
these forms are given below. 

34. The scandicus (from the Latin word for 'ascend') : the three 
notes ascend and are of equal length or duration. This neume seldom 
carries any rhythmic sign. 

456 245 

Fig. 20 

35. The quilisma (from the Greek word for 'turn'): this neume 
also consists of three notes that ascend, the first of which, however, 
is prolonged. The quilisma is distinguished from the scandicus in 
that the middle note is 'wavy.' It is probable that this note was 
formerly trilled; possibly it is parent to our modern trill or turn 
sign. One fact concerning the quilisma, however, is certain: the 
note preceding the wavy one is always to be prolonged. The second 
and third notes of this neume are rendered smoothly and, compared 
with the first, rather briefly. 


6 7 1 

Fig. 21 

The quilisma exerts a retroactive influence on the notes approach- 
ing it in such a way that the closer they are to it, the longer are they 
held. The note immediately before the quilisma, however, is never 
as long as two beats. In this way the following beautiful melodic 
figure of the Requiem Introit is given due prominence and grandeur. 
As the notes before the quilisma approach it, they grow longer in 
deference to it. 



■ "< ♦ , 

lu- ce- at 

Fig. 22 



36. The salicus (from the Latin for 'leap') : in this neume the 
second note is always prolonged. We may readily see the reason for 
its name, for the first note represents the brief 'leap* upon the 
syllable; the second note, being long, represents a slight settling 
upon the syllable; and the third, the brief leap from the syllable. 
This neume is distinguished from the scandicus in that it always 
bears the rhythmic ictus over or under its second note. 


456 673 
Fig. 23 

4 1 2 

37. The climacus (from the Latin for 'ladder') : in this neume all 
three notes descend. The first always has the virga; the others are 
inclined notes (See par. 11 [b]). 


654 643 

Fig. 24 

38. The torcidus (from the Latin word pertaining to 'a press' — 
a meaning exemplified by the shape of the neume). In this neume 
the middle note is highest and is to be sung smoothly. 



23 2 

Fig. 25 

1 43 

39. The porrectus (from the Latin for 'extension') : this neume 
consists of three notes, the middle of which is lowest. It is the re- 
verse of the torculus although its construction differs from it. The 
first note is at the left of the curved bar; the second is at the right 
of the curved bar ; and the third is obviously a square note. It should 
be noted that the first and second notes of this neume are incor- 
porated within the sweeping black stroke of the neume. 



656 167 

Fig. 26 



40. The tristopha (from the Greek for 'triple comma') : a neume 
of three notes, all of them in unison. In order to keep the choir 
together on these notes, and to add a delightful nuance, each note 
is to be sung with a very slight percussive attack. 



111 666 

Fig. 27 

41. In former editions of the chant, the so-called i oriscus J con- 
sisted of a slightly inverted square note affixed to the last note of a 
neume, more frequently, to the last note of a torculus. In more 
modern editions its place is taken by an ordinary square note. 
According to Dom Mocquereau (Le Nombre Musical Gregorien II, 
pg. 377, par. 499) it changes neither the rhythm nor the melody of the 
neume to which it is added. It does, however, have a retroactive 
effect, that is, it informs the chanter that the foregoing neume is to 
be rendered with utmost delicacy and smoothness. Since it is no 
longer printed as an oriscus and since it effects neither the rhythm 
nor the melody, it is excluded from the present classification of the 

in ancient editions 

in modern editions 


Fig. 28 

42. The compound neumes are either enlargements or combina- 
tions of simple neumes. They can be easily recognized if the simple 
neumes are known. There is, however, a peculiarity in the forma- 
tion of the neume compounded with the torculus. When a note is 
added to the torculus the figure looks as if it were a porrectus with 
a note before it. 

6 6 


5 5 

4 5 

Fig. 29 


And so. too. when two notes are added to it: 

6 2 i 

6 : *<: 

5 5 16 

Fig. 30 

43. Under special neumes are enumerated the so-called 'liqui- 
scents' (meaning 'smooth-flowing'). They are distinguished from 
ordinary neumes in that the last note is printed smaller than the 
preceding note or notes. They do not differ from the ordinary 
neume in melody or rhythm. Their sole purpose is to facilitate and 
beautify the singing. In chanting we pass from the vowel sound of 
one syllable to the vowel sound of the next, briskly but clearly 
inserting the consonants that occur between them. Thus, for ex- 
ample, in singing Do - mi - nits, the vowel sound of the syllables 
o. z, and u are sustained but the consonants d, m, n and 5 are quickly 
sounded in their place before or after the vowel. But between the 
principle vowels of the syllables there may be and often are other 
vowels. For example, the three principle vowels in the word exaudi, 
from the Requiem Introit. are e. a and i. The sol over the second 
syllable is sung on the -a; la is printed as a liquescent on which -ir- 
is sung. 

5 — — \ 

• — ■— = = — * — » 

printed ~ j sun^ ■ * 

exaudi ex - a - u - di 

Fig. 31 

Among the consonants that appear between the vowel sounds, there 
are some that are singable and sustainable. These are called 'semi- 
vowels' because they are so much like vowels in their production. 
A vowel is a sound that allows the air to pass freely through the 
mouth without the interference of the tongue, teeth or lips. A semi- 
vowel also allows the air to pass freely but because there is a slight 
interference of the tongue, lips or teeth, or any combination of them, 
they can be sung to a definite note and sustained. The semi-vowels 
are /. m, n, r and ng. When these semi-vowels appear at the end of a 
syllable, that syllable is sustained on the principle vowel and the 
semi-vowel is sung to the liquescent note, as was illustrated in the 
case of the vowel u in the word exaudi. In other words, the semi- 
vowel is rendered as though it formed a diphthong with the principle 
vowel. The soft, pleasant variety of the delicate shade of the 
semi-vowels in contrast to the lengthier tone of the actual vowel 



IT 43 

sounds adds a beautiful nuance to the singing (see par. 253). The 
following illustration gives the ordinary and the liquescent form of 
the various neumes: 





* ( I ) i 






Ch.) % 

Fig. 32 

The following illustration shows the proper method of singing the 
words landa, Sanctus and Alleluia, which are generally written with 
liquescent neumes. The singer must bear in mind, however, that 
the whole of the extra vowel or semi-vowel is not sung to the 
liquescent, for this note is merely a help for its proper rendition. 




tau -da — la - u - da 

San-ctus — Sa - n - ctus 




Al - le - lu - ia 

A - I - le - lu - ia 

Fig. 33 


If 44-46 


44. The contrast between long and short notes in the chant is so 
important, as we shall see in succeeding chapters, that a special 
chapter will be devoted to an explanation of their difference and 
appearance in print and to the methods by which the long notes are 

45. In paragraph 20 we dealt with the interrelation of melody, 
rhythm and text. A note, as we see it in Gregorian notation, has a 
definite rhythmic character, a definite pitch, and is to be sung to 
a definite syllable of the text. Every note in the chant is, therefore, 
influenced by these three features. As a rhythmic entity it is either 
a rise note or a fall note; as a melodic entity it is either high or low. 
Furthermore, it is sung to a syllable of a word which is either a 
grave or an acute accented syllable. For this reason it is quite in- 
correct to say that all notes are exactly equal in time value even 
though they may appear to be equal in print. Every rhythmic rise 
note is slightly briefer than a rhythmic fall note precisely because 
of its rhythmic character; and, on the other hand, the note on the 
accented syllable of the word will always be slightly briefer than 
that on the last syllable because the accented syllable is briefer in 
speech than the other. In this way the features of melody, rhythm 
and text will influence the length of a note. 

46. How is one to determine the time value of an ordinary note 
in Gregorian notation? By an ordinary note we mean one which is 
not to be prolonged. When Gregorian melodies are transcribed into 
modern notation, the ordinary note is represented by the modern 

eighth note ( j* ). A long note may be represented by a quarter 

note ( J ), or by an extended eighth ( J^ ). This, however, indicates 

only the relative value. What is the length of the ordinary note in 
Gregorian Chant? Each ordinary note has the time value of a syl- 
lable spoken in ordinary discourse. This is what the square note 
originally intended to convey. (See par. 11 [c]). To find the tempo of 
a piece of Gregorian music we need only recite the words of the 
text in a devout fashion; the time value given to each syllable will 
be the time value of the ordinary note. The chant should never be 
either dragged or rushed. It must be remembered that every Gre- 
gorian selection is a prayer — whether of adoration, praise, suppli- 
cation or petition. No one should attempt to sing the chant without 
the firm realization that the song is a prayer, a prayer which must 



spring from a lively faith and profound love toward Him whom we 
address in song. The living faith will then rid the chant of all 
sluggishness, but the profound love will check all speed. 

47. The secret of the simple beauty of the chant is hidden within 
the steady and uninterrupted pulsation of each note. The time 
value of the ordinary note is called the 'basic pulsation'. The basic 
pulsation must be maintained evenly throughout the piece. This 
observation may, at first hand, appear superfluous, but it has been 
noted that more errors in singing are committed on this point than 
on any other. Whether the notes are high or low in pitch, their basic 
pulsation must be maintained constantly. Frequently in the more 
elaborate chants we may find as many as thirty or forty notes on 
one syllable, as for instance in the following excerpt from the Re- 
quiem Gradual, yet, in the same piece some syllables are given but 
one note: 

H*r^ a ■■ ■ •! — i : ■ l— 

ae-ter- na 

Fig. 34 

Nevertheless, whether there are many notes or only one note to 
the syllable, the basic pulsation of each ordinary note is the same. 
There must be no slackening of speed on the syllables with single 
notes, nor acceleration on those of many notes, for the basic pul- 
sation is even throughout the selection. 

48. Every note, then, is of equal duration except those that are 
marked long. The following signs of length are used: (1) the dot; 
(2) the episema (par. 16); (3) repetition; (4) the pressus; (5) specific 
notes in some neumes. 

49. The dotted note: a dot added to a note doubles its value; 
that is, it prolongs its time value to two beats. Within a division 
a dotted note has the value of two ordinary beats. At the cadence, 
however, where greater length is indicated, the dotted note will 
be prolonged to two slow beats. 

fi— 37" 

Fig. 35 

The word 'cadence' is derived from the Latin word cadere meaning 
'to fall'. By a cadence, therefore, we mean a fall in rhythm, as that 
which occurs at the end of the divisions (see par. 12.) As the division 
is of greater or less importance, the cadence will accordingly demand 


If 50-51 


a greater or less retard or slackening of tempo; the incise will be 
accorded a very slight cadence or retard, the member with a more 
pronounced retard, but the phrase with a very definite slackening 
of tempo. This will be explained more fully in a subsequent chapter. 

50. The episema (par. 16): a horizontal line over or under one 
or more notes. This mark does not necessarily prolong the note to 
two beats but merely indicates that the note or notes should be 
extended beyond their ordinary value. (For further discussion see 
par. 145.) 








Fig. 36 

The episema over or under the note of a podatus, however, is ai 
special case (see 3rd and 4th neume in Fig. 36). An episema under 
the neume indicates that only the first or lower note is to be pro- 
longed; the sign over the neume tells that both notes are to be 
prolonged beyond their natural value. 

5 1 . Repetition : a note may be prolonged by the addition of 
other notes in unison to it. The addition may be made to a single 
note, as in the following example from the Requiem Gradual: 


■ ■ 

ab audi- 

Fig. 37 

or to the first or last note of a neume, as the sign over Re- in Re- 
quiem, from the Requiem Introit, and the sign over -ne of Domine 
from the Gradual: 



■ ■ 

Re-qui- em 




mi- ne: 

Fig. 38 

This kind of prolongation is used in active passages while the dotted 
note is used in rather restful passages of the rhythm. 



II 52 

52. The pressus: in order to procure certain rhythmic effects 
two neumes are often so fused that the last note of one and the 
first note of the next form one long note. As examples we may refer 
to the pressus (clivis and porrectus) over -ter- from the Requiem 




ae -ter - nam 

Fig. 39 

over -la (podatus and clivis; two clives), and over the ju- (climacus 
and clivis) of the Gradual: 


M *** W 






t a I 

■ a 


i 'f 








. -• 




■ ■ 





In the Tract of the Requiem Mass there are three pressi over 
■ctonim of dejunctorum and four over the -i of perfrni: 

C ■ ' ?Tl fl^'ft 


de- functo - rum 

^ ' ^,, f h h < .""V\^' ^i^. 

per-fru- i. 

Fig. 41 

Some authors do not distinguish between a repetition and a pressus. 

Such a distinction should be made because of the different rhythmic 

effect. A pressus always implies a slight 'pressure' or stress on its 

'first note; that is, on the first note of length. The stress is light if 

!the figure occurs on a weak or unaccented syllable. The pressus on 

-ter- (Fig. 39) is definitely stressed because -ter- is the accented 

i syllable of the word; that on -la (Fig 40), however, is slightly 

^stressed since this is the last and weak syllable of the word. 


1T 53-54 


53. Specific notes in some neumes: there are two neumes in 
which one note is always prolonged : the first note of a quilisma 
(par. 35) and the second note of a salicus (par. 36). There are two 
salici and two quilisma in the following examples from the Requiem 

latr. J- 





*imm Ji> 


t r 



equi- cm * ae-tcr- nam do-na e- is 




M O V. ' iqy^tfc^*^ j [ 



ne: et lux perpe-tu- a lu- ce- at 

is. Ps. Te de- 

Fig. 42 

54. A good example of the frequent occurrence of long notes is 
found in the Introit of the Requiem Mass given in the above illus- 
tration. The note over Re- of Requiem is long because of repetition; 
the second note over ae- of aeternam and over do- of dona are long 
because they are the second notes of salici ; the first sol over -is of 
eis and over -at of luceat are long since they are the first notes of 
quilisma neumes. The seven dotted notes are clearly shown. In 
this short selection there are four of the five indications of length. 
There is no example of the episema, however. In the Offertory of 
the Requiem Mass (exclusive of the Versicle), to which the student 
may refer, we have: 23 dotted notes; 5 notes or figures with episema; 
4 examples of repetition and 7 quilisma. 


1. The contrast between brief and long notes is very important in 
the chant (par. 44). 

2. The length of notes is slightly influenced by the text and the 
rhythm (par. 45). 

3. An ordinary Gregorian note has the time value of an ordinary 
syllable in Latin prose (par. 46). 

4. The time value of an ordinary note is called the 'basic pulsation' 
of that melody (par. 47). 

5. All notes are practically equal in duration except those marked 
long (par. 48). 

6. The marks of length are: (1) the dotted note, (2) the episema, 
(3) repetition of the same note over the same syllable, (4) the 
pressus and (5) the first note of the quilisma and the second note 
of a salicus (pars. 49-53). 


J 55-58 


55. In general, rhythm is a movement; specifically, it is a flowing 
movement. The word 'rhythm' is derived from the Greek word 
rhein which means 'to flow'. Anything that rises and falls in its 
progress through space, like the waves of the ocean, the flight of a 
bird, the bouncing of a ball or even the lifting and setting down of 
the foot in a walk, may rightfully be called a rhythmic movement 
through space. 

56. Because of the similarity of the tones in a melody to the rise 
and fall in nature, the word is also applied to the movement of music. 
Every melody is like a flowing body that tends toward a climax 
and seeks rest after the climax is reached. But such movement in 
music does not take place through space but in time. In Gregorian 
chant we are concerned exclusively with melody produced by the 
voice whose rhythm consists solely in the rise and fall of the tones 
as they rise towards a climax and fall to rest from it. 

57. If, then, a melody is to have rhythm or, in other words, is 
to be set into a flowing motion, there must be a constant rising or 
falling in each of its notes. But how is this effect produced in the 
tones? There are three means especially that create a flowing move- 
ment within the melody. If one note is high and the other low in 
pitch, the one seems to rise and the other to fall, for the rise in any 
movement is always high and the fall, lower. Then, again, if one 
note is light and the other, heavy or stressed, the light note leaves 
the impression that the note is rising and the heavy, that the note 
is falling, for a rising body is always relatively light and a falling 
body, always relatively heavy. Finally, if one note is brief and the 
other long, the brief note seems to rise and the other to fall, since 
a flowing body never remains long in a suspended position but rises 
only to fall again. These concepts, therefore, are the three features 
that produce a rhythmic movement: pitch, stress and duration. 
The first concerns scale tones: the highness or lowness of a note; 
the second, concerns weight: the lightness or heaviness of a note; 
and the third, its extent in time: the brevity or length of the note. 
We shall now examine the character and importance of each. 

58. Pitch. When a note which is high in pitch is followed by one 
that is lower, there is conveyed a definite feeling of motion. If we 
sing 2 1 2 1 2 1, we notice how the rhythm rises on 2 and falls on 1. 
But we saw (par. 20) that melody and rhythm are interdependent. 
Can we, therefore, produce a real rhythmic flow when the rise note 



is lower in pitch than the fall note? We can if we add another feature 
to counteract the inversion of pitch. If we sing 12 12 12, making 2 
definitely longer than 1, we see how the lower note seems to rise 
and the higher but longer note seems to fall. The reason lies in the 
concept that brevity indicates a rise; length, a fall. Melody, then, 
is very important, but not the most important, feature of a rhyth- 
mic movement; if we can invert it, it must not of necessity be essen- 
tial to the rhythm. Furthermore, we can sense a definite rhythm 
in the beating of a drum although it is devoid of all melody. There 
is even a distinct rhythm in the dot and dash of telegraphy. Hence 
we conclude that, since rhythm can be created without melody, 
melody is not an indispensable feature of rhythm. 

59. Stress. In regard to stress, as was stated above, notes also 
seem to rise and fall when one is light and the other heavy or 
stressed. A rise note will, of course, always be light, for a body must 
yield its weight in order to rise. But must a fall note be heavy? We 
know from experience that when a body falls, it will touch the 
ground with a stress proportionate to its weight. A rubber ball falls 
much more lightly than a billiard ball. A ball of yarn falls even more 
lightly than either of these balls because it is lighter in weight. A 
snow flake falls to the ground without the least perceptible stress. 
But a musical tone is even lighter than a snow flake. Its fall must 
therefore be lighter. Indeed, it is immeasurably light, for the snow 
flake is after all a material thing of material weight. But no measur- 
ing device, however delicate, can determine the weight of a tone. 
In singing, the tone is made by the breath and lives on the air; the 
sound we hear is air set in vibration. Modern music bases its rhythm 
on stress, on the heavily accented beat. But this is contrary to the 
spirit of the chant. Stress would, indeed, render the rhythm of the 
chant very clumsy and would steal from it its peaceful flow. Chant 
melodies as such are without stress; when stress does appear, it 
comes to them not from the rhythm but from the nature of the text, 
or the heaviness of the syllable on which a note is sung (see Chap. 
VIII). From this consideration we deduce a practical hint for sing- 
ing the chant more beautifully: the rise notes should be sung lightly 
and briefly and the fall notes will take care of themselves. We may 
apply here the same principle which holds in the material world: 
an object will fall to the ground of itself but it must be raised by a 
force outside itself. When we throw a ball into the air, we need not 
be concerned with its return to the ground. This it will do of itself. 
Likewise this will be true in singing the chant. The rise notes should 
be lifted with brevity and the fall notes will fall of themselves. The 
unwarranted concern about the nature of the fall note lies only in 
our confusing the chant rhythm with that of modern music. When 
in the future the fall note is referred to as 'weighty', it should be 
remembered that 'weighty' is used (a) as a weight proportionate to 
the weight of a musical tone and (b) as a correlative to the lightness 
of the rise note. 

60. Duration. By duration we mean the contrast between brief 
and long notes. This is the most essential feature of a rhythmic 



movement in the chant. The rise note must of necessity be brief 
and the fall note long or at least comparatively longer than its rise. 
We have already observed the influence of duration (par. 58) where 
it produced a real rhythmic flow even when the melody was inverted. 
The underlying reason for the importance of duration lies in this: 
that a body always remains at rest until it is brought into motion. 
If it is raised, it falls to rest as soon as possible after its rise. Further- 
more, a moving body rises only to fall again. No one lifts the foot 
in walking for the purpose of keeping it suspended, but in order to 
lower it at a more distant point. When the step is completed the 
foot rests at the fall, never at the rise. We can invert melody, we 
can dispense with stress, but we cannot invert or dispense with 
duration and still procure a real rhythmic movement. Duration 
is the essential feature of a rhythmic movement. A rise can never 
be long and a fall can never be brief in any natural movement. 

61. What happens if we make the rises long and the falls short? 
We shall have a sort of movement which a lame person executes 
when he walks. In the case of a person suffering from a severe pain 
in the foot, if he walks at all, his foot will remain suspended as long 
as possible. He will lower it only to renew support and then for the 
briefest possible moment. In this instance the rises (lifting of the 
foot) are long, and the falls (lowering of it) are brief. This rhythmic 
flow is neither natural nor beautiful. The nature of rhythm is 
distorted: the rises are long and the falls brief. In music this kind 
of rhythm is one which modern musicians call 'syncopation'. There 
is no place in the chant for such unnatural rhythm. In Gregorian 
music every rise is brief and every fall is long. 

62. We have found that the ordinary notes of the chant are 
either square or diamond-shaped and that the most common note 
is the square punctum. But regardless of the shape of the note, all 
are practically of equal duration unless there is some special indi- 
cation of length. Yet when we see two notes that are exactly alike 
in print, for example, two square notes, and when we know from 
the grouping that the first is a rhythmic rise and the other a rhyth- 
mic fall, then the fall note should always be slightly longer than the 
rise because the essential quality of the fall note is length. That is the 
reason why every ordinary note has practically the same duration. 
The monks of Solesmes (to whom we owe the recent restoration of 
the chant), intended to mark some of the notes with a'c! meaning 
'celeriter' or 'quickly', when they first published their rhythmic edi- 
tions. But because of the objections raised by those who disagreed 
with them, the authorities at Rome thought it best to exclude these 
signs. The fact remains, however, that the rise notes are always 
brief and the fall notes comparatively long. This subject will be 
treated further in succeeding chapters. 



Elements of 


procured by notes 


high in melody, light in stress, brief in duration 


low in melody, weighty in stress, long in duration 





63. Rhythm, we have seen, is a movement that flows in a series 
of rises and falls. We have observed that the essential quality of 
the rise is brevity and that of the fall is length. We have learned, 
in other words, that the rise of the rhythm, whether it be high or 
low in pitch, must be brief in duration, and that the fall, whether 
high or low in pitch, must be comparatively long in duration. 

64. Because every rhythmic movement must consist of at 
least one rise and one fall and because the rise is essentially brief 
and the fall essentially long, w r e should be able now (if the theory 
is correct) to procure a true and complete rhythmic movement 
through the agency of a brief and a long note. The brief note 
represents the rise to action and the long note the fall to rest. 

} j 

R F 

Fig. 43 

65. This combination of a brief rise and a long fall constitutes 
in reality a complete rhythmic movement. It has beginning and end, 
a rise to action and a fall to rest. One without the other is meaning- 
less and incomplete, but united, the two form an entity which is a 
complete rhythmic movement. This combination is known as an 
elementary rhythm. It is elementary because it contains only the 
elements of a rhythm; it cannot be further divided without destroy- 
ing its nature; it is a complete rhythm because it consists of a rise 
and a fall. 

66. We must study this combination more thoroughly, for in 
grasping it clearly we shall learn the elements of every rhythmic 
movement. As a chemist by analyzing one test-tube of water from 
a lake learns the ingredients of all the water in the lake, so we, by 
analyzing the smallest possible rhythm, learn the elements of every 
rhythmic movement. 

67. The brief rise note expresses activity and liveliness; it is 
the beginning of a movement, the transition from the state of rest 
to the state of motion. The rise note, then, being brief, is the active, 
lively portion of the rhythmic movement. 



68. The fall note, on the other hand, being long, expresses repose 
and completeness. It is the end of a movement, the return to the 
state of rest from the state of motion. The fall note, being long, 
expresses rest and repose and constitutes the peaceful portion of 
the rhythmic movement. 

69. We can see how natural and reasonable this theory is by 
comparing the rise and fall notes to a step in walking or a word in 
speech. In taking a step we lift the foot and immediately put it 
down. The lifting is brief, corresponding to the rise of a rhythm; 
the setting is long, corresponding to the fall of the rhythm. The 
former is naturally brief because we lift the foot only to put it down 
again; the latter is long because the step is completed and ended 
at that point. In pronouncing a word, for instance, A-men, we notice 
that the first syllable is brief and the last, long. One syllable is 
comparatively brief because it is the beginning of the word; the 
other is long because it is the end: 

j> j 

£ \ 

A - men 
R F 

Fig. 44 

It is evident that a step would cease to be such if we inverted the 
order, making the lift long and the setting brief. The foot would 
remain suspended; anyone witnessing this unnatural movement 
would grow anxious to know when the act would be completed in 
a long setting. This is true also with the word. The word A-men 
would be as unnaturally distorted as the step if we rendered the 
first syllable long and the last, brief: 

j j> 

A - men 
F R 

Fig. 45 

Furthermore the flow of rhythm of the word is smoothly delineated 
by the joining of the accent marks, the acute and the grave: 

A - men 
R F 

Fig. 46 


This discussion reiterates and further clarifies the principle: that 
every rise is brief and every fall is long. 

70. Melodies are not composed of a series of elementary rhythms 
any more than the ocean is considered a composite of waves. In 
order that the elementary rhythm become a part of the whole, it 
must lose itself, as it were, within the whole, just as one wave loses 
itself in the swell of the sea. The rise of each elementary rhythm is, 
and always will be, brief; but the fall note, while remaining a fall, 
must sacrifice some of its finality in becoming a part of the whole. 
Again a comparison with walking may clarify the subject. In taking 
a single step the foot is lifted and immediately put down : the lift is 
brief and the support long because it constitutes the end of the step. 
But in taking a walk of four steps, each of the first three supports, 
while remaining supports, are not prolonged because the walk is not 
(but terminated at any of these points. If each support 
were long, we would hardly consider the series as one walk but rather 
as four individual steps. In order, then, to unite them into one new 
element, into a walk, the falls yield some of their length to become 
a part of the whole. There is no doubt that these supports are not 
places of permanent rest but rather places of temporary support. In 
the following figure we show a series of elementary rhythms which, 
having long falls, do not form a complete entity: 


Fig. 47 

Each fall implies a separate end of the preceding rise. But in Figure 
48 the falls, while remaining falls (but falls of support), sacrifice 
part of their being to become a part of the whole. 

R F R F R F R 

Fig. 48 

71. This element may be compared to a word in speech as was 
done in the elementary rhythm. In pronouncing the word De-us, 
the last syllable will be slightly prolonged because it forms the end 
of the word. Likewise, if we pronounce the words Pa-ter and Je-su 
and Chri-sti individually, the last syllable of each will be prolonged 
because it constitutes the end of each word. But when we join the 
four words into one continuous phrase, De-us Pa-ter Je-su Chri-sti, 
the last syllables, while remaining the ends of the individual words, 



yield some of their length in order to become a part of a new phrase. 
It is unnatural and incorrect, therefore, to pronounce the four words 
individually as 


■ ' 
















Fig. 49 

for in this form they still remain four separate individual words. 
But when the last syllables give up part of their length while re- 
maining last syllables of words, they form a new entity, one whole 
phrase. The last syllable of the last word will, however, be long 
because this syllable is not only the end of the word but the end of 
the whole phrase as well: 

■ ■ 



■ 1 ■ ■ ' 

De - us 

Pa - ter 

Je - su Chri - sti 


.„*' ^ ~* \.. 

De - us 

Pa - ter 

Te - su Chri - sti 

Fig. 50 

We observe how readily the flow is delineated if we join the direc- 
tion of the grammatical accents. We should remember too, for it 
will serve us to good purpose later, that the so-called 'acute* accent 
is delineated in an upward movement and the 'grave' accent, in a 
downward movement. 

72. The rhythmic falls of the chant take place on the ictus notes 
(par. 15). The word is derived from the Latin meaning 'thrust' or 
'tap'. But we must not be misled into the belief that the Romans 
'thrust' or 'struck' these notes as though they were the stressed or 
heavy or accented beats, like those of modern music (par. 59). On 
the contrary, the falls in Roman and Greek music were so delicate 
that an external aid was needed to convey the rhythm to the dancers 
and to secure uniformity of motion. This aid was supplied by the 
'tap' made by the director; hence the use of the word 'ictus'. The 
ictus, then, is that note on which the rhythm dips down for support 
during its flow or for rest at the end of the flow. Compared to the 
flight of a bird, it is that note on which the bird, after opening its 
wings, closes them to lift itself higher or, on its downward flight, 
closes them to prevent a sudden drop. If compared with a step in 
a walk, they are the settings of the foot, with this difference, how- 
ever: in walking the foot is set with a weight proportionate to the 
weight of the body, but in the chant the settings are weightless 
because the flowing element, the melody, is entirely without weight. 



As has been said above, too much stress should not be laid upon the 
correct rendition of the ictus notes; after all, if the rise notes are 
lifted through brevity and lightness, the falls or supports or ictus 
notes will take care of themselves. 

73. In future lessons there will be employed a number of terms 
to designate this rhythmic beat. It might be called a 'touch,' a 
'support', a 'fall' or an 'ictus' note, for all these terms are synony- 
mous. In observing the movements of a bouncing ball we learn an 
important aspect of the rhythm of the chant. When a ball bounces, 
the rise before a fall and the rise after that fall merge and become 
one at the moment of support. In the same way the ictus note is the 
end of one rise and the beginning of the next. Thus it serves as a 
unifier of the notes. In point of fact, it is the great unifying force 
among all the notes of the chant. The rise note is always brief, the 
fall or ictus, comparatively long; the one is active, the other restful. 



Fig. 51 

74. From the above two kinds of rhythmic falls may be dis- 
tinguished: a fall of support and a fall of rest, one being a fall to a 
temporary rest and the other to a permanent rest. The first is long in 
comparison with the rise; the second is long in reality. In our modern 
conception of rhythm too much emphasis is laid upon the correct ren- 
dition of the heavy or accented beats of a measure. Such a procedure 
may be warranted in modern music, but in the chant we heartily 
recommend that the rise of the brief note be given our special atten- 
tion. The rise note is the active portion of the rhythmic movement 
and its activity lies in the brevity with which it is rendered. In this 
sense the fall notes of support will be comparatively longer. These 
falls are always long by nature for two reasons: (1) they descend to 
the place of rest although they do so only for support; and (2) the 
rhythmic movement can be ended at any of these points but never on 
a brief rise note. In practice, therefore, to procure the feel of the 
rhythm and to insure the rhythmic flow, it is well at times definitely 
to prolong the falls and to abbreviate the rises; whereupon the 
notes can be equalized and the rhythmic flow just experienced will 
be maintained. The greater the contrast of duration, the more 
definite is the rhythmic flow: 



jv J.. J> J.. $ J.. 

J> J. J> J. J> J. 



J> J> J> J> J» J> 

Fig. 52 

75. The rhythmic notes may be considered in various combina- 

(1) An elementary rhythm is made up of a brief rise and a long fall 
of completion. 

(2) A simple rhythm consists of a brief rise and a fall of support. 

(3) A 'time group' or simply a 'group' of notes consists of an ictus 
or fall of support with one or two rise notes. This last com- 
bination is not a rhythmic group since a rhythm consists of 
a rise and a fall in that order and not the contrary. For this 
reason it is not called a 'rhythmic group', but a 'time group' 
or simply a 'group' of notes. 

R F R F F R F R R 

elementary simple time group time group 

rhythm rhythm of 2 notes of 3 notes 

Fig. 53 


1. A rise note is essentially brief; a fall note essentially long (par. 

2. The .combination of a brief note representing a rhythmic rise 
and of a long note representing a rhythmic fall constitutes a real 
rhythmic movement (par. 64). 

3. The above combination is called an elementary rhythm (par. 65). 

4. A knowledge of the essences of an elementary rhythm reveals a 
clear knowledge of all rhythm (par. 66). 

5. The brief rise expresses activity; the long fall expresses repose 
(pars. 67, 68). 

6. The length of the fall of an elementary rhythm is sacrificed for 
the unity of the whole (par. 70). 



7. The rhythmic falls of the chant take place on the so-called 'ictus' 
notes (par. 72). 

8. There are two kinds of fall notes: (1) a fall of support which is 
long in comparison to its rise and (2) a fall of rest (either tem- 
porary or permanent) which is long in reality and generally 
marked as a dotted note (par. 74). 

9. Notes can be combined in various ways : 

(1) A brief rise and a long fall is called an elementary rhythm. 

(2) A brief rise and a fall of support is called a simple rhythm. 

(3) A fall of support and one or two rise notes is called a group of 
notes: a group of two composed of a fall and a rise, and a 
group of three composed of a fall and two rise notes. These 
are called 'time groups' or simply 'groups' (par. 75). 


H 76-78 



76. Rhythm is the soul of a melody. Its presence endows the 
composition with life and unity: with life in so far as the melody is 
in a constant state of motion within itself; with unity in so far as 
the notes are gathered into one whole, sacrificing their individuality 
and becoming parts of a great movement. What the soul is to the 
body, rhythm is to the melody. Just as the soul gives life to the 
body and keeps it from disintegrating, so rhythm gives life to the 
melody and converts the collection of the many notes into one vital 
flowing melody. Moreover, no note may ever be deprived of its 
rhythmic character of a rise or a fall note because as soon as the 
rhythm ceases, then also will there be a cessation of vitality and 
unity. The life of a melody consists in its rhythm ; rhythm consists 
of a brief rise and a long fall. 

77. When all the notes are gathered into one grand rise and fall 
movement, each note sacrifices its individuality to become a part 
of the whole. The note will never lose its identity as a rise or fall 
note even though it yields its individuality, just as a drop of water 
is lost in a wave although it never ceases to be what it is. Similarly, 
words that make up a poem remain individual words even though 
they have relinquished their individual existence in favor of the 
whole. When the notes of a melody are unified and vivified through 
the agency of rhythm, we hear not a series of individual notes but 
one sweeping melody made up of many notes. We recognize therein, 
although we may not necessarily know how or why, the delightful 
pleasure of order, the beauty of composition and the splendor of 
created life. Where there are life and unity, there also are peace 
and beauty. 

78. In preceding chapters we have analyzed a rhythm to learn 
its essential elements. From now on we shall concern ourselves with 
the task of synthesizing the notes; that is, unifying them into a 
sweeping whole. This is possible only if we know the essence of a 
rhythmic movement. If all the notes are marshalled under one rise, 
which must be brief, and one fall, which must be long, then we have 
succeeded in converting a series of many notes into one new entity: 
a living, unified whole, one rhythmic movement. In Chapter V we 
fused two notes into a simple or elementary rhythm. In Chapters 
VII and VIII we shall combine the notes into groups. In Chapter IX 
the groups will be united into musical divisions and in Chapter X 
the divisions will be united into complete musical phrases, the larg- 
est of all divisions. These are the means whereby the purpose of 
rhythm is achieved. 




1. The purpose of rhythm is to form a complete whole of all the 
notes of a melody (par. 76). 

2. When notes are united into one whole the selection is endowed 
with life and unity (par. 77). 

3. Our present task consists in uniting all the notes of a selection 
in one complete whole (par. 78). 





79. Every note of a melody is either a rhythmic rise or a rhyth- 
mic fall note and as such is a member of a group of notes. 

80. Every group of notes is made up of an ictus or fall note (pars. 
72, 75) and one or two rise notes. While the ictus note is naturally 
the end of every rhythmic movement, it is considered the first of 
every time group. In other words, the ictus or fall note forms the 
unit according to which notes are grouped. This procedure is both 
practical and theoretical : it is practical because the ictus is the out- 
standing note of a rhythm and the note at which the rise notes are 
fused (par. 73) ; it is theoretical because the ictus is in reality the 
end of a rhythmic movement though considered here as the be- 
ginning of a group. 

81. No ictus note may ever follow another ictus note directly. 
This is reasonable for a moving (or flowing) body must rise after 
each fall in order to prepare for another fall. 

82. One rise note must follow a fall or ictus note; two rise notes 
may follow an ictus note. A group which is made up of one ictus and 
one rise note is called a 'time group of two notes' or simply a 'group 
of two'; a group which is made up of one fall note and two rise 
notes is called a 'group of three' notes : 

1 2 

1 2 3 

■ ■ ■■- 

■ ■ 

time group 
of two notes 

■ ■ ■ 

time group 
of three notes 

Fig. 54 

83. In modern music the rhythm is regular in the sense that the 

rhythmic flow descends for support at regular intervals. In modern 

2 4. 

— or even — time, the rhythm supports itself regularly in 

f . 3 3 

groups of two's; in modern -Tror — time the rhythm falls regu- 

o 4 

larly in groups of three's. In the chant, on the other hand, the 

rhythm is free in that it descends for support in groups of both 

two's or three's. 




Modern music 

j j , j j 






f J ' J J \ J J d \ J d J - \ « H| 

Gregorian rhythm 


1 2 

Fig. 55 

There is no set standard for the sequence of groups since the rhythm 
of the chant is born of and built upon the freedom of Latin prose 
in which groups of two's and three's follow each other indiscrim- 

84. In grouping notes in a Gregorian melody, then, (1) every 
note belongs to a group of notes of which the ictus is always the first; 

(2) no ictus may follow immediately one upon another ictus; 

(3) no ictus may carry more than two rise notes. The grouping of 
notes in modern music is clearly expressed by the time signature, 
bar lines and the time value of the notes. The grouping in the chant, 
however, is determined according to fixed principles. 

85. The grouping of notes in neumatic chants is established by 
the neumes. The law for neumatic grouping may be expressed as 
follows: the first note of every neume is the ictus. There are two 
exceptions to this law: (1) when the second note is definitely marked 
as the ictus (par. 15); (2) when the second note is long. 

86. In regard to the first exception, the second note in a neume 
may be marked as the ictus in order to produce a certain rhythmic 
effect. In this case the first note loses the ictus quality. 









Fig. 56 

If the third note of a neume, however, is marked, the first note 
may still retain the ictus quality. 


*. tt 

Fig. 57 

In regard to the second exception that the ictus is not on the 
first note of a neume if the second note is long, there are two consid- 
erations: (1) the second note is always long (and always marked with 
the ictus) in a salicus (par. 36) ; (2) the first note loses the ictus when 
the second is fused into a long note by the pressus (par. 52). These 
are illustrated in the following figure: 





separate neumes 


1*— f* — N3r 

12 12 123 12 12 123 




E ft 


(2)12 12 



Fig. 58 


IT 87 

Ictus and length are so closely associated that the first note of all 
length is always the ictus note except when the second note is 
marked with the ictus (this happens rarely) : 

g * = 


1 23 

1 23 


1 2 


1 23 



1 2 

exception : 




Fig. 59 

In the case of the pressus, three conditions must be present: (1) the 
last note of one neume and the first of the next must be in unison ; 
(2) they are printed close together because (3) they are to be sung on 
the same syllable. For example, in Figure 42, we can observe how 
the separate neumes are spaced when they are distinct and how close 
together they are when the notes form a pressus. 

87. We have seen that no note may ever stand alone or that every 
note belongs to a group. But what happens to notes that are separ- 
ated or isolated from their neumes or that stand alone between 







Fig. 60 

Such notes become members of the preceding group. If the pre- 
ceding group is made up of two notes, the isolated note recedes and 
becomes the third of that group (Fig. 61 [a]); if, however, the pre- 
ceding group is made up of three notes, the isolated note divides the 
four into two groups of two's (Fig. 61 [b]) : 




12 3 12 

12 3 123 

Fig. 61 

I 88-89 




12 1 2 

1 2 


12 1 2 

1 2 

Fig. 61 (cont.) 

88. Sometimes a melody or new phrase begins with a single 
note. Such a note cannot stand alone and yet there is no preceding 
group to which it can attach itself. In the case of a new phrase the 
isolated note cannot recede to the last group of the previous phrase 
since the phrase mark indicates completion (par. 12 [b]). In such 
cases a rest must be supplied. This rest functions as a silent ictus. 





anctus, * Sanctus, 
7 2 1212 3 12 12 

Fig. 62 

The use of the rest is practical inasmuch as a melody or phrase may 
begin with either a rise or a fall note. Very often when the text 
begins with an accented syllable, the chant delights in setting that 
syllable to a rise note in order to give it its proper lightness. This 
nuance in turn lends an active beginning to the phrase. 

89. When an incise or a member begins with an isolated note, 
that note follows the same rule of recession as described above (par. 
87), since neither incise nor member implies completeness (12 [c], 
12 [d]). The following sections of the Requiem Offertory illustrate 
the foregoing rule: 



m am 

■ i > 

■ ■ ■ 

■ ■ 


3 V 

■ ■ 

3 It ■ 


B f 

■ * ?!♦ « l 

2 12 3 12 


12 12 12 12 

3 12 3 12 12 12 



N s % 

■ ■ * 

|3 P 

■ 9 ■ 

12 12 12 

12 12 

Fig. 63 




90. An analysis of the grouping of the first phrase of the Requiem 
Introit is given in Fig. 64. The letters above the staff refer to the 
rhythmic groups of two and groups of three. Notes marked with 
arrows indicate isolated notes which recede, becoming members of 
the preceding groups. 


a b 

f f 



i! N ^: I J f* 

12 12 1 

2 1231212 1 2 12 

3 12 12 



g d g d 

gg d ff 




12121 2 12 12 

Fig. 64 

a First note of length 

b Four notes divided into two groups of two's 

c Last syllable of word (see next chapter) 

d Marked as ictus 

e First note of length (pressus) 

f Dotted notes 

g First notes of neumes 


1. A group of notes is made up of an ictus and one or two rise notes 
(par. 80). 

2. No ictus note may follow another ictus note directly (par. -81). 

3. Rules for Grouping Notes in Neumatic Chants (par. 84 ff) : 

(a) A note is the ictus or first of a group if it is so marked (par. 85). 

(b) The first note of every neume is the ictus unless the second 
note is marked or long. 

(c) An isolated note recedes and becomes a member of the pre- 
ceding group. If the preceding group is composed of two 
notes, the isolated note becomes the third; if the group is 
made up of three notes, the series is divided into two groups of 
two each; if it is made up of more than three notes, the ictus 
will be placed on a virga (if it occurs) or on an important 
note of the mode. 

(d) If the isolated note begins a new phrase or a new melody 
(in which case it cannot recede to a preceding group), a silent 
ictus is supplied as the first note of that group; the isolated 
note becomes the second note of the group and the group is 
counted rest 2. 




91. In syllabic chants there are few, if any, neumes (par. 19 [a]). 
Just as the grouping of the neumatic chants is determined by the 
neume (par. 85), so the scheme of grouping for syllabic chants is 
based upon the nature of the syllables of the text. Our present task 
is to learn which syllables will most naturally support the rhythmic 
ictus. This note, as in the neumatic grouping, will be the first of 
every group. The same principles apply now as before: no ictus may 
succeed another immediately ; every group is composed of one ictus 
and one or two rise notes (par. 85). 

92. In order, then, to render the chant in its original purity, we 
must learn to respect the Latin word as it was pronounced in the 
days when the melodies were composed. The character of the melody 
and the flow of the rhythm (pars. 20-23) are to a great extent based 
upon the nature of the words in the text. 

93. Syllables are unified into words through the agency of 
rhythm; that is, through a rise and fall movement. We know that 
a rise and fall movement may be procured by means of pitch, stress 
and duration (par. 57). In Cicero's time (about the beginning of the 
Christian era) duration was the governing principle, especially 
among the cultured. But among the unlettered the principle of 
melody was maintained only until about the 11th century. From 
that time on, stress gradually took the place of melody. This last 
change dealt the death blow to the Latin language as a popular 
language and gave birth to the modern Romance languages of Ital- 
ian, French and Spanish. It is to be remembered that the syllables 
were equal in duration and were rhythmized according to the agency 
of melody during the golden age of the chant. 

94. When melody was the principle of rhythmizing syllables 
into words, all syllables were largely of equal duration. The rise and 
fall were procured by a gradually ascending pitch on the syllables 
preceding the accent and a gradual descent in pitch on the syllables 
following the accent. Although the feature of melody thus became 
pre-eminent, still it was impossible to drop entirely the other two 
features of rhythm. The gradual rise in pitch was accompanied by 
a gradual increase in volume (crescendo) and the descent in pitch 
was accompanied by a decrease in volume (decrescendo). Further- 
more, the feature of duration was retained, although in the form 



If 95-96 

of a slight hastening of tempo (accelerando) before the accent, and 
a slight slackening of tempo after it (rallentando). 

De - us 
Ma-rf -a 
om - nf- po - tens 
om - ni - po - ten - tis 
jus - ti - fi - ca - ti - 6 - nes 

Fig. 65 

95. *The accented syllable, as can readily be seen, is without 
doubt the most important syllable of the word. It forms the turning 
point of the rhythmic movement by determining the end of the rise 
and the beginning of the fall movement. It is the climax, the highest 
part of the word, and seems to draw the syllables to itself (by the 
increase of volume and tempo) and to restrain those after it (by 
the decrease in these elements). In this way the syllables of every 
word are unified and vivified into one new being. 

96. In the correct rhythmization of the chant the accented syl- 
lable is very important and yet little understood today. First, there 
is not the least implication of stress in the original meaning of the 
word 'accent'. It is derived from the Latin words ad (to), and cantus 
(sung), and implies something melodious. The words were later 
contracted to accantus, and still later changed to accentus. Today 
when we speak of an 'accented' syllable or an 'accented' beat, we 
always think of something 'stressed' or 'emphasized by pressure*. 
This concept formed no part of the original meaning of the word 
'accent' and must form no part of our interpretation of the chant. 
It must be remembered, therefore, that the syllable was, and for us 
who desire to sing the chant correctly still is, the 'sung' or 'melo- 
odious' syllable of the word. Secondly, it is pre-eminently the 
high syllable of the word since it forms the end of the rise and the 
beginning of the fall in the rhythm of the word. Although it is 
essentially high in melody, it partakes nevertheless of the other 
qualities of the rise; namely, it is also light and brief. At about the 
11th century this syllable became lengthened and stressed; this 
treatment, as was said, was one of the factors that gave birth to 
the Romance languages. The opposite of the 'accent' or high syllable 
is naturally the low or 'grave' syllable, from the Latin gravis (heavy) 
whence came the French 'grave meaning 'low'. But when the ac- 
cented (the adcantus) syllable became stressed, it was called the 
'acute-accent' syllable. And so today we speak of the 'acute' and 
the 'grave' accents. Here, however, is a confusion of terms: a syl- 
lable cannot be both grave and accented since 'grave' implies 'low- 
ness', and 'accent', 'highness'. Henceforth, therefore, in order to 



keep before the student this basic difference, and at the same time 
to be more accurate in regard to the chant, the terms 'accented' and 
'grave' will be employed only in reference to syllables. The accented 
syllable, often termed the 'tonic' or 'primary' syllable, can also be 
regarded as the rise syllable, the grave as the low or fall syllable of 
the word. 

97. In regard to the position of the accent in Latin, all words 
divide into two kinds: spondee and dactyl. The spondee is a word 
whose accented syllable is next to the last syllable : 

de - pre - ca - ti - 6 - nes 

The dactyl is one whose accented syllable is the second from the last : 

Do - mi - nus 
om - nf- po - tens 
ju - sti - fi - ca - ti - 6 - ni - bus 

98. In addition to the primary accent, longer words also have 
secondary accents. These are found on every other syllable, count- 
ing back from the primary accent, in both spondees and dactyls: 

p = primary accent 

s = secondary accent 





s p 

4 3 2 1 



s p 
6 5 4 3 2 



- per 

ad - ju - van - dum 

- ve - re - an - tur 

- ex - al - te - mus 


ad - ju - to - ri 

ma - gni - fi - cen - ti 

- pre - ca - ti - 6 - ni 

- um 

- a 

- bus 

99. Now we are prepared to analyze the nature of syllables to 
determine on what syllables the rhythmic ictus can most conven- 
iently fall. In view of paragraphs 69 and 71(q.v.), the most natural 
place for the ictus is the last syllable of words. There are four reasons : 
(1) the last syllable, being the end of the word, is most naturally 
low, long and weighty, having the same characteristics as the ictus 
note (par. 72) ; (2) it is most natural that the end of the word should 
correspond with the end of a rhythm rather than the beginning; 
(3) the contrary treatment, that is, the agreement (in quality) of 
the last syllable with a rise note which is naturally high, light and 
brief (par. 96) would violate the very nature of the word ; (4) the last 
syllable will naturally be a final rest if this is the last word of the 
text. The last syllable of a word, moreover, will be a fall because it 
is the end of a word; the fall will be qualified, however, and will 
be less long when the word is not the last in the text. 

100. Since rhythm has a law of its own apart from the text 
(pars. 20 - 23), it does not happen invariably that the rhythmic 
ictus or fall note coincides with the last syllable of words. But since 



the rhythm should respect the nature of the word as far as possible, 
we do find that most words have the ictus on the last syllable. It is 
quite evident that when the last syllable of a word is set to a rise 
note of the rhythm (that is, when it is pronounced in such a way 
that the last syllable is high, light and brief), it cannot be styled 
'natural'. Consequently, a word has its natural rhythm when the 
last syllable coincides with the ictus or fall note; if not, it is deprived 
of its natural rhythm. This rule has been attacked on the ground 
that if the ictus is on the last syllables of the text, those last syl- 
lables are 'accented' or 'stressed' — a treatment which mutilates 
the Latin word. But those who propose this objection fail to realize 
that the ictus note is not a stressed note; it is rather the fall of a 
rise note and as such it need not necessarily be heavy (par. 59). 
In the Offertory of the Requiem Mass there are 52 words of more 
than one syllable. Of these, four are definitely deprived of their 
natural rhythm. 1 

101. In rhythmizing chants, if the ictus is placed on the last 
syllable whenever possible, some ictus will be made to support 
more than two notes. Syllables other than the last syllables must 
be found to carry the ictus. 

102. If we allow a spondee its natural rhythm by placing an 
ictus on the last syllable, the accented syllable cannot have an 
ictus, for two ictus notes cannot stand side bv side: 




- ter 

re - cor 


- re 


- Stl 

- fi - ca - ti 
Fig. 66 


- nes 

Hence the accent of a spondee generally occurs on the rise note of 
the rhythm. In other words, in a spondee the accent and ictus 
alternate. Sometimes, however, the chant will set a neume on 
the accent. In this case the ictus and accent coincide: 

I ■ ■ ■ 

om - ni - po - ten - tis 
Fig. 67 

J The four words are: libera (first), fidelium, prof undo, morte. (See Fig. 79.) 



103. If we allow a dactyl its natural rhythm the word accents 
can also carry the ictus: 

glo - ri - a 
ab - sor - be - at 
ju - sti - fi - ca - ti - 6 - ni - bus 

Fig. 68 

As if to allow both rhythm and text their independence, the chant 
will sometimes place a neume between the accented syllable and 
the last syllable in a dactyl, thus removing the ictus from the accent: 

t . 

■ ■ ■■■Li* 
i • [■ i 

de - pre - ca - ti - 6 - ni - bus 

Fig. 69 

This is particularly true of the special dactylic cadences which are 
explained in the chapter on Psalmody (par. 212). 

104. We have noted that the ictus is placed on alternate syl- 
lables, counted back from the last known ictus. But we know also 
that the secondary accent is placed on alternate syllables, counted 
back from the primary (or last) accent (par. 98). The rhythm will, 
therefore, always be the same: 

Fig. 70 

In the following figure we shall see what happens to the different 
accents when different types of words are set to this rhythm : 

ju - stf - fi - ca - ti - o - nes 
ju - stf - fi - ca - ti - d - ni - bus 

Fig. 71 

In spondees the ictus and accent alternate; that is, the ictus will 
occur on grave syllables. In other words, the accents of spondees 
occur on rise notes, the accents of dactyls occur on fall (or ictus) 



105. What is the importance of this consideration? Herein lies 
one of the most charming features of Gregorian rhythm and one 
of the fundamental principles of the theory of Solesmes. In modern 
music the word accent is made to fall always on the heavy beat of 
the rhythm. The monks of Solesmes state that rhythm and text 
are independent of each other; that the ictus may or may not occur 
on the accented syllable. The importance of this principle lies in 
this fact: the accent of the word is always high, light and brief 
(par. 96) ; the fall of the rhythm is always low, long and weighty 
(par. 68). Definite phenomena appear when the accent occurs on 
the rise and when it occurs on the fall note. 

106. When the accent of the word occurs on the rise note of 
the rhythm, we have concurrence of two like forces; the lightness of 
the accented syllable of the word and the lightness of the rise note 
of the rhythm. Because the most important feature of the rise is 
brevity, such a note must be sung with particular brevity; it 
must be brief both because of the rise note of the rhythm and be- 
cause of the accent of the word. We may compare this rendition to 
the effect of a ball thrown upward by a smart toss of the hand ; the 
hand represents the rhythmic rise and the ball represents the word 
accent. Both rise together but the quicker the toss of the hand, the 
higher the ball will rise. This movement can be pictured by joining 
the accent marks of the syllables and by indicating the crescendo 
and decrescendo to and from the accent: 


\ /' *\ /' "\ accent and grave syllables 

V * 


rises and falls of rhythm 

om - nf - po - ten - tis 

Fig. 72 
In such a rhythm the flow leaps over the accents of the words. 

107. When the fall of the rhythm occurs on the accent of the 
word, we have a contention of two opposing forces; the accent of 
the word will always be high, light and brief, while the fall of the 
rhythm must be low, long and weighty (par. 68). How can these 
two opposing forces be reconciled on an identical note? The answer 
is: by stressing the accented syllable. This and the pressus are the 
only cases where the fall note of the chant is heavy or stressed (par. 
59). The reason for the stress is that the accent of the word must be 
lifted. The example of the hand and the ball can illustrate the effect. 
The hand again represents the rhythm and the ball, the accent. If 
the ball is thrown downward with definite pressure, it will rebound 


II 108-109 


upward from the floor. The more forceful the downward thrust, the 
higher will be the rebound. In this way both elements are satisfied. 
We may note the progress in this delineation, observing that there 
is a double action on the same note: 


^n accent and grave syllables 

■ ? n ? ■ 

., rises and falls of rhythm 

ju - sti - fi - ca - ti - 6 - ni - bus 

Fig. 73 
In such a rhythm the flow leaps upon the accents of the word. 

108. It is well to treat the secondary accents of the words in 
the same way as the primary accents whenever possible; that is, 
if the primary accent is on the rise note, the secondary accent or 
accents should be set to rises when possible. This will insure the 
similarity of accents in the same word. 

109. This peculiarity of the ictus and accent, in one case co- 
inciding and in another alternating with each other, lends a definite 
charm to the chant. While ordinary rise notes are high, light and 
brief, those rise notes set to the accents of words are especially high, 
light and brief or, rather, higher, lighter, and briefer than the others. 
In the same way an ordinary fall is low, long and weighty, but a fall 
on the accent is lower, longer and weightier than an ordinary 7 fall. 
In other words, the accents of words magnify the qualities of the 
rhythm. We cannot fail to see how the composers have diligently 
tried to vary the accents in all of the chant selections, at times even 
achieving a perfect balance within the selection. In examining the 
Pange lingua (Mode III) we note how the accents of one division 
are on the ictus while those of the others are on the rise or off the 
ictus. We shall consider secondary accents as we do the primary 
accents (which indeed they would be if the words were shorter) ; 
monosyllabic words will be compared with longer words of a corre- 
sponding line of the hymn. There are twenty accents: 



f 110-111 




■ ■ 





J i. 


ange lingua glo-ri- 6-si Corpo- ris myste-ri- um, Sanguinis- 




!• T 

■ ■ 

• ■ 

■ 7 

1 ■ — ■ r 

que pre-ti- 6-si, Quern in mundi pre- ti- um Fructus ventris ge-ne- 




ro-si Rex effu-dit gen-ti- um. 

Fig. 74 
Accents on rise notes (12) 

Pan - 

ge lin 

- gua 

San - 

gui - nis 

- que pre - ti - 

o - si 

Fru - 

ctus ven 

- tris ge - ne - 

ro - si 


ef - fu 

- dit 


on fall notes : (9) 

glo - 

ri - 6 , 

- si 

Cor - 

po - ris 

mys - te - ri 

- um 


in *mun 

- di pre - ti 

- um 

gen - 

ti - um. 

In the first line two accents coincide with the ictus, two alternate; 
in the second line all accents coincide with the ictus; in the third 
line, all accents alternate; in the fourth line, all accents coincide 
with the ictus; in the fifth line, all accents alternate; in the sixth 
line, one accent coincides with the ictus and the other accent is off 
the ictus. This peculiarity can hardly be laid to chance; it was un- 
doubtedly the wish of the composer to vary the accents. 

110. Two places have already been found for the ictus in syl- 
labic chants: (1) on the last syllable of words; (2) on the accents 
of dactyls. In rare cases when these principles do not suffice or when 
there may be a choice, the ictus will be preferably placed on mel- 
odically low notes rather than on high notes since the ictus is natur- 
ally low and long. 

111. Thus far we have considered the natural rhythm of the 
word. The melody, however, also affects the word (par. 93). It is the 
nature of the Latin word to be pronounced with a slight rise in 
pitch to the accented syllable and a slight descent in pitch from it 



(par. 93). When the melody therefore respects the word in such a 
way that it places on the accented syllable a note higher than that 
of the last syllable, we say that word has its natural melody. It 
may happen at times, however, that the melody rises and in so 
doing it will ascend even if higher notes are set to last syllables. 
For example, in the Requiem Mass the melody treats the first two 
Sanctus differently from the third. On the third the melody climbs 
in such a way that the last syllable is higher than the accented 
syllable : 


1 1 

V ■ 


■ V 

■ ■ ■ 



1 1 

anctus, * Sanctus, Sanctus D6mi-nus 

Fig. 75 

The first two Sanctus are treated naturally in respect to the rhythm 
and melody, but the third is deprived both of its natural melody 
and its natural rhythm: of the natural rhythm, because it lacks an 
ictus on the last syllable; and of the natural melody, because the 
last syllable is higher than the accented syllable. 

112. What practical value has this consideration? When a word 
is deprived of its natural melody, it is often well to deprive it of 
its natural rhythm. This is advisable because in this case the accent 
becomes stressed and it seems to lend the proper downward pressure 
needed before an ascent. Before a person can jump up from the 
floor, he must prepare himself for the upward spring by a downward 
pressure. This pressure is supplied by the heavy accent. Of course 
the voice is different from the body; it needs no spring before an 
ascent. On the other hand, the chant is so calm and natural that it 
requires no more of the voice than nature demands of the body. 
In another example, the beginning of the Salve Regina: 




(a) ■ B 

■ • 
- — r 

: (b) 

■ ■ 


Sal - ve Re - gi - 


Sal - ve Re - gi - 


Fig. 76 

we observe how clumsily the rhythm rises at (a) in which we give 
the word Salve its natural rhythm by setting an ictus on the last 
syllable. But at (b) the ictus is set to the accent which renders the 
accent stressed. In the first figure the word has its natural rhythm; 
in the second figure it is deprived of its natural rhythm because it 
is deprived of its natural melody. 




113. At times it will be either impossible or inadvisable to give 
each word its natural rhythm. In such cases, since the last syllable 
has more claim to the ictus than the accent, words should be given 
their natural rhythm, whenever possible, by setting an ictus on the 
last syllable: 

Two dactyls : no - mi - ne Do - mi - ni 

rather than: 

no - mi - ne Do - mi - ni 

Fig. 77 

Yet the rhythm of the chant is sufficiently free to allow the director 
a choice. For example, in the last verse of the Dies irae, the words 
Pie Jesu, Domine, dona eis requiem, could be rhythmized as at (a) 
below : 




T— ♦ 


Pi- e Je-su Domine, dona e- is requi- em. 


7 ■ 

T ■ 

T ■ 


Pi- e Je-su Domine, dona e- is requi- em. 

Fig. 78 

In this way each word is given its natural rhythm. The rendition 
at (b), however, seems preferable for these reasons: 1) the heavy 
accents on the words Pie Jesu Domine lend a definite seriousness to 
the last prayer of the long sequence; 2) these accents, being heavy, 
form a delightful contrast to the many light accents throughout 
the selection, especially those of the first and like verses; 3) if the 
words Pie Jesu were given their natural rhythm, the dactyl Domine 
could not support an ictus on the accent although it is almost needed 
for ascent on the last syllable (par. 107). However, as was said, either 
choice is equally correct; the very freedom of the chant gives to the 
individual director the freedom of choosing what appears to him 
best in each case. 




114. As an exercise in finding which words have their natural 
rhythm and which their natural melody, we can examine all the 
words of the Requiem Offertory: 

Offert. -g 




■*- ft 


N^Hi a fife I * j 

o-mi-ne Jesu Christe, * Rex glo- 
* . * 

n- ae, 


!■■ J 

■ ■ 

■ ■ 

■ft= ft 


be-ra a-nimas 6m-ni-um fi-de- li- um de- functo- rum de poe- 

s j~ { . — = r ~ 

rr 1 

"* ■ ■ J*"i ■* ' ■ " B A ■* 

■ *m m 

nis infer- ni, et de pro-fundo la- cu: li-be-ra e- as de o- re 

< ' 11% /» 


-a »■ 

fl- 1 

le- 6- nis, ne absor-be- at e- as tar- ta-rus, ne cadant in 

*^a fg^i- i P a ■• g /■ a . ^ 


-a — ■- 

obscu- rum: sed signi- fer sanctus Mi-cha- el repraesentet 


a: j fttP p ]0 ^ 

■« — ft 


a • 

fl- 1 

e- as in lu- cem sanctam: * Quam o-lim Abrahae promi- 



8 a nJ H i tffr 

■ ■ 

J 8 P ft 



si-sti, et se- 

mi- ni e- jus. ^. Hosti- as et 
** ** , 

pre-ces ti-bi Domi- ne laudis of-fe-rimus: tu susci-pe pro 



U 114 




Pb A 



m.* «• 

i ■ • 

- i .. . 



a i 



V ■• 

1 i 






ho-di- e memo- 

ri- am fa-cimus: fac 




• • 


r ' fr ■"■ 

■* ■* 

■ * ■ a 

P« a 

1 ■ 



e- as, D6mi-ne, de mor- te transi- re ad vi-tam. * Quam o-lim. 

Fig. 79 

*■ Words having natural melody but lacking natural rhythm 
** Words having natural rhythm but lacking natural melody 

There is a total of 52 words having more than one syllable; 4 of 
these have their natural melody but their unnatural rhythm, and 
10 have their natural rhythm but unnatural melody. The remaining 
38 are treated as natural from the point of view of both rhythm 
and melody. 


1. The rhythm of syllabic chant is based on the nature of the 
syllables of the text (par. 91). 

2. The accented syllable is the highest part of the word (par. 95). 

3. Spondees have the accent on the next to the last syllable; 
dactyls have it on the second syllable from the last (par. 97). 

4. The most natural place for the rhythmic ictus is on the last 
syllable of a word (par. 99). 

5. A word whose last syllable carries the ictus is said to have 
natural rhythm (par. 100). 

6. Spondees cannot support an ictus on the accented syllable 
(par. 102). Dactyls generally can (par. 103). 

7. When the accent of a word occurs on the rise of the rhythm, 
that accent is especially high, light and brief (par. 106). 

8. When the accent of a word occurs on the fall of the rhythm, 
that accent is especially low, long and weighty (par. 107). 

9. A word is said to have its natural melody when the accented 
syllable carries a higher melodic note than the last syllable 
(par. 111). 

10. In cases of doubt, and where these rules are not sufficient to 
supply the place for the ictus, it is well to remember that the 
ictus is preferable on low rather than on high notes (par. 113). 

11. Whenever possible, the secondary accents of words are to be 
treated in the same way as the primary accent (par. 107). 


If 115-119 


115. In the two preceding chapters we united individual notes 
in neumatic and syllabic chants into groups consisting of an ictus 
and one or two rise notes. The process of unification will now extend 
to the forming of the groups into rhythmic movements. 

116. Just as every note is either a rise or a fall note (par. 79), 
so every group is either a rise or a fall group. 1 

117. The first note of every group is the ictus note. The ictus 
note is a rhythmic fall either of support or of rest (par. 73). This 
note may fall for support in order either to rise higher or to enjoy 
a temporary rest. The flight of the bird illustrates these elements. 
A bird whose destination is a high mountain peak will soar up to 
the peak not in a straight line but in a series of curves necessitated 
by the opening and closing of its wings. Each closing will bring the 
bird higher, but each closing will also necessitate a reopening. At 
the reopenings, therefore, the bird will lose altitude and will descend 
somewhat until the next closing. Thus, the closing of the wings 
corresponds to a rise group in rhythm, the reopening to a fall group. 
On descending from its destination the bird will repeat the same 
process with this difference however: the openings will be the more 
important part of the flight; the wings will be closed only to break 
the force of gravity and to make the fall more smooth. 

118. The rise group has, therefore, the qualities of the rise note: 
it is high in melody, light in stress, and especially brief in duration. 
The fall group, correspondingly, possesses the qualities of the fall 
note: this is low in melody, weighty in stress, and especially long 
in duration. The term 'weighty in stress' is used in relation to 
'lightness in stress' of the rise note. Although the rise group is 
indeed light, the fall group is not arbitrarily "heavy", only rela- 
tively "heavier" than the rise group. The rise group is the active, 
lively group while the fall group is its restful counterpart. 

119. Which groups will, therefore, be rendered as rise groups 
and which will be rendered as fall groups? The following is the gen- 
eral rule: after the syllable of the word is satisfied, the melody de- 
termines the nature of the groups. Whenever possible, 2 the accented 

■In many texts on chant rise and fall are termed respectively arsis and thesis. 
For the sake of practical clarity the author prefers the terms rise and fall group. 

2 The qualifying condition, "whenever possible," is suggested for on occasion 
the melody disregards the nature of the syllable since text and melody are not 
governed by the same principles. When, therefore, an important climax occurs on 
the last syllable of a word, that particular group of the climax will be rendered as 
a rise regardless of the nature of the syllable. Notice the important climax on the 
last syllable of eis and luceat in the Requiem Introit. 



If 119 

syllables will be rendered as, rise groups and the grave syllables as 
fall groups. This is done to satisfy the nature of the syllable (par. 94) . 
But when the nature of the syllable has been satisfied, the grouping 
will follow the flow of the melody in such a way that ascending 
melodies will be rendered as rise groups and descending melodies 
as fall groups. Finally, the first group on the primary accent of a 
word will always be rendered as a rise*and the last syllable as a fall — 
for the rise group is to the rise note what the fall group is to the fall 
note (par. 117). The following is an analysis of examples from the 
Requiem Mass: 

(a) Kyrie: 



1 2 
R R 


5 6 




V i m ■ ■ • 

y- n- e 




Fig. 80 


Groups 1 and 2 : rise groups - 

of word. 
Groups 3 and 4: fall groups 

Group 5 : rise group - 
Group 6: fall group - 
Group 7: rise group 

slightly stressed). 
Group 8: fall group - 

rising melodies on accent 
- falling melodies on last 

a secondary accent of the word. 

a grave syllable of the word. 

- the primary accent (this ictus is 

the last syllable of the word. 

(b) Gradual: 

12 3 4 5 6 7 8 9101112 




r^ ^ fU 


mi- ne : 

Fig. 81 

Groups 1 and 2: rise groups — on accent of the word. 
Group 3: fall group — the grave syllable of the word. 
Group 4: fall group — the last syllable (from here the 

melody rules). 
Groups 5, 6, 8 and 9: rise groups — the melody ascends. 
Groups 7, 10, 11 and 12: fall groups — the melody de- 


% 120-121 


120. Further analysis of the nature of groups will finally estab- 
lish the general principle that one group grows out of the other. One 
rise group which follows another will be a more marked rise than 
the first and will be slightly more active; likewise, one fall group 
following another fall will be a more marked fall and will be slightly 
more restful. Furthermore, whenever a fall group follows a rise 
group, that fall will constitute a slight repose from the preceding 
rise. An example may serve to make the principle and its applica- 
tion clearer. When we ascend a mountain we may sit down along 
the way for temporary rest. We sit down not at the foot of the 
mountain but at various points in our ascent. And so upon descend- 
ing, we rise from the place where we rested. Similarly, in regard to 
the groups, the rise group which follows a rise is slightly more pro- 
nounced because we are nearer the climax. A fall that follows a rise 
grows out of that rise; that is, it is a slight repose from the rise out 
of which it has grown. In summary, the rhythm of the chant is 
never abrupt: one group calmly and smoothly grows out of the 

121. One rhythmic group grows more active or becomes more 
reposeful than the preceding through the aid of rhythmic features 
known as dynamics and agogics. By dynamics is meant the gradual 
increase or decrease of volume or of loudness and softness. This is 
indicated in modern music (and can also be applied here) by means 
of the opening and closing dynamic angle: 



Fig. 82 

The dynamic angle opens to the highest note on a primary accent of 
the word. Each rise group that precedes this accent grows, as it 
were, toward it and each group that follows it fades away into silence 
from it. In the Requiem Mass, for example, the melody treats the 
word Kyrie most naturally: 

12 3 4 
m R R FF 


Ky- n- e 

Fig. S3 



Group 2 is slightly louder than group 1 ; group 4 is slightly softer 
than group 3; group 3 is slightly softer and more reposeful than 
group 2. By agogics is meant the gradual increase or decrease of 
tempo designated in modern music by the words accelerando and 
rallentando respectively. In the chant this takes the form of a very 
slight increase in tempo before the accent and a very slight decrease 
after it. The reason for the change in tempo lies in the fact that 
rhythm is a movement. There is a rise to the climax (the accent of 
the word in the above example) and a fall from it to a state of rest. 
The rise seems to convey the feeling of a gentle anxiety to reach 
the destination (the accent) and the fall, the calm enjoyment of 
having reached it. In this way the word is rhythmized by its groups. 
The rhythmizing by groups can be compared with the example in 
par. 119. Each word is given, whenever possible, its proper rise to 
the accent and fall from it. A further example from the Requiem 
Mass can be found in the following: 

Tract. J- 
VIII. — 


■ a 




bsol- ve, 

* Do- mi-ne, 

f 5 5 67 
1 2 




1 1 





mi - 






Fig. 84 


1. Every group is either a rise or a fall group (par. 116). 

2. The word and the melody determine the nature of the group in 
this way: 

(a) Accented or rise syllables will, as far as possible, be accorded 
rise groups (par. 118). 

(b) Grave or fall syllables will, as far as possible, be accorded 
fall groups (par. 118). 

(c) After the nature of the syllable has been respected, the 
melody will rule in such a way that ascending groups will 
be rendered as rises and descending groups, as falls (par. 119). 

3. Dotted notes will, in the majority of cases, be rendered as a fall 
group since the fall group corresponds to the fall note (par. 116). 

fNOTE: the notation in the above illustration and in the following chapter 
have been transcribed into numerals so that the student may have a more graphic 
picture of the rhythm. 


If 122-125 



122. Chapters VII and VIII presented the combining of indi- 
vidual notes into groups. In the preceding chapter (IX) the groups 
were organized into their rise and fall movements according to their 
position on the syllables of words. In this chapter the process of 
unifying the notes will be extended by unifying the words into musi- 
cal divisions. Incises will be united into members and members into 
phrases. Since the phrase is a complete entity in itself (par. 12), the 
process of unification must end with the unifying of the phrase. 

123. Each note must always retain its character as a rise or a 
fall note and each group must likewise preserve its qualities as a 
rise or a fall group. Consequently each division will preserve its 
character of rise or fall. However, each of these simple or compound 
parts will contribute its share to the rise and fall movement of the 
whole phrase. As each word in a poem expresses its own idea and 
yet loses itself in the context, so each note, each group, each sub- 
ordinate division (incise and member), retains its individuality but 
is lost within the structure of the whole phrase. And just as a note 
sacrifices its length to become a part of the greater whole (par. 70), 
so each subordinate division sacrifices its length to become a part 
of the phrase. 

124. Each subordinate division of a phrase has its own rise and 
fall movement. The rhythm will rise to a climax in that division 
and fall to rest from it. The climax of a division is generally the 
highest ictus note on an accented syllable. When the melody ignores 
the nature of the word, the climax may be set to the last syllable 
of a word (see the two phrases of the Requiem Introit). But in the 
more ornate selections, the climax is strictly melodic. In these cases 
the particular kind of notes used will readily reveal the climax 
(par. 32). For example, in the Requiem Gradual, the bivirga over 
the last syllables of Domine, perpetna, and mala indicate without 
doubt the melodic climaxes of their respective melodies. 

125. An analysis of the first phrase of the Requiem Tract will 
serve to clarify the theory of unifying members and incises into 
phrases. The phrase, the text of which is Absolve, Domine, animas 
omnium fidelium defunctorum is divided : 

, . , {1st incise: Absolve 

1st member < ~ , . ^ 

(2nd incise: Domine 

^ , , /3rd incise: animas omnium fidelium 

2nd member ... • • , , . J 

\4th incise: defunctorum 




1f 126-127 

126. The first incise consists of one word whose accent is on the 
syllable -sol-. Both melody and rhythm respect the nature of the 
word in that both rise to the accent and fall to rest from it. The 
notes are grouped as follows: 

Tract. £- 
VIII. — 


■ a 


bsol- ve. 

5 5 6 7 






1 2 




R R 




Fig. 85 

Considered apart from the whole, this incise will rise from silence 
and fall to rest in silence. The climax is found in Group 2, the ictus 
of which is stressed slightly because the ictus and the accent of the 
word coincide (par. 107). Groups 1 and 2 are rendered actively as 
rise groups and the others reposefully as fall groups. Groups 3 and 
4 become increasingly more reposeful (par. 120) to Group 5 which 
is the end of our present division. Such repose or falling to rest is 
expressed by two elements of rhythm, namely, slowness in time and 
softness in volume. Groups 3, 4 and 5 grow slower and softer in such 
a way that Group 5 vanishes away slowly and softly into silence. It 
should be noted that the syllables -sol- and -ve begin on the rise 
notes of their groups: 

Tract. I 





Fig. 86 

If we extract these syllables from their context, we find that the 
-sol- and the -ve are accorded a simple rhythm since the second note 
of each is an ictus, the first being a rise note (par. 75). This binds the 
syllables of the word very closely for each syllable begins on the rise 
or last note of the preceding group. We should note also how the 
melody falls from la to ti la on the syllable -sol- and from la to 
la sol on the syllable -ve. 

127. The second incise consists of the word Domine whose 
accent is on the first syllable. Here again both the rhythm and the 




melody respect the nature of the word. The notes are grouped as 
follows (again for purposes of analysis each incise is considered 
distinct from the whole and as an entity in itself) : 




1 1 






mi - 














Fig. 87 

A rest is supplied for the ictus on Group 5 because this note is 
isolated by the on-coming pressus (par. 88). This word, considered 
apart from the whole, begins at silence and falls to rest into silence. 
Groups 5, 6 and 7 are active rise groups ; groups 8, 9 and 10 are restful 
falls. The syllable Do- begins on a simple rhythm. The syllable -mi- 
is rendered on one rise note. This is quite brief and should not be 
prolonged unduly. 

128. The first member then consists of the two first incises, 
Absolve, Domine. The end of the first incise will sacrifice its length 
and its return to silence in order to become part of the whole. When 
considered apart as a plain incise, it ended in a long fall to silence. 
Now, however, it neither descends into silence nor is its end as long 
as before ; here it is but a temporary fall for support in preparation 
for the rise on the next incise. The first sol of the Domine has become 
a part of the last dotted sol of Absolve (par. 87). The two incises 
could not be more intimately bound together, for the first note of 
Domine is really a part of the last group of Absolve. A group of three 
is formed by the dotted sol and the isolated sol, the new syllable 
beginning on this note (see par. 138). 


VIII. — i" 



bsol- ve, 

Do- mi - ne, 

Fig. 88 


The climax of the whole member is the re of the accented syllable 
Do-. The climax of the first incise now drops down to the level of a 
secondary accent of the member. Since the ictus note re of Group 7 
is the climax, it receives a slightly stronger stress than the ictus of 
Group 2 which has now become a secondary accent. 

129. The third incise is syllabic in character. It contains only 
two neumes (two podatus) : one on the accent of animas, the other 
on the accent of fidelium. Since the accent on fidelium is the climax 
of the incise, it will receive a slightly stronger stress than that of 
animas. The incise is grouped as follows: 


a-ni-mas omni- um fi- de- li -um 

5 1 1 

111 11 

6 1 1 

1 . 

a- ni- 

mas omni- um fi- 

de- li - 



12 13 




F U* F 



Fig. 89 

Here are three dactyls, each possessing its natural rhythm (ictus 
on the last syllable, par. 99). But a variety of accents is offered: 
the words animas and fidelium have neumes on the accents and must 
therefore be stressed, but omnium has its accent on the rise note of 
Group 12. The majesty of this incise is displayed by the three groups 
of three's (Groups 11, 12 and 14). Group 11 is clearly an active rise 
group; Groups 12 and 13 are fall groups of support in preparation 
for the climactic accent on Group 14 whose ictus is definitely stressed 
There must be a constant, almost eager, rise to this accent. Further- 
more, each note is of equal value except the last ; for this reason there 
may be no tarrying or hurrying on any syllable. 

130. The fourth incise consists of the word defunctorum whose 
accent is the syllable -cto-. The melody and rhythm respect the 
nature of the word. In grouping we must remember that the first 
note of length is always an ictus, for pressus abound in this short 
incise. The fourth incise is grouped as follows: 

* U stands for undulation (see par. 133). 


f 131-132 


■ ■ 





de-fun-cto- rum 


7 23 

• • 

1 1 


• • 
1 1 

6 i 


• • 

1 1 

6 i 



defun ■ 

■ cto- 






















Fig. 90 

The climax of this incise is undoubtedly Group 17 on the accented 
syllable. Group 16 is a fall for reasons explained in par. 139; Group 
17 is the climax because of the high melody and its position on the 
accented syllable. Groups 18 and 19 are falls, the first because of 
the lower melody, the second because of the last syllable. Group 20 
is preferably rendered as a fall because of its length and because it 
prevents too high a rise on the secondary climax of Group 22. But 
Group 21 is a rise preparing for the secondary melodic accent of 
Group 22. The succeeding groups are falls, each helping to bring the 
phrase to a peaceful end. 

131. The second member consists cf two incises, one of which 
is syllabic, the other melismatic (par. 19). The primary climax of 
the member is Group 17 ; Group 14 of the first incise is now relegated 
to secondary importance. The member begins actively on a stressed 
ictus, descends for a support on Groups 12 and 13 to prepare for 
the secondary climax on Group 14 whence it falls for another support 
on Group 15. Group 16 is preferably rendered as a fall of support to 
prepare for the grand climax on Group 17. The rhythm now falls 
towards the end of the word but rises quietly again on Groups 21 
and 22. These are but secondary rises which break the monotony 
of the falls. In uniting the two incises, Group 15 is not long and does 
not return to silence: it is simply a fall of support preparing for 
the accent on Group 17. (See Fig. 91 on the insert.) 

132. The whole phrase is, therefore, made up of four incises 
which are condensed into two members. Each incise and each mem- 
ber is distinct from the other by its own rise and fall movement. 
This distinction is brought out by brief rises and long falls which 
distinction is, as was stated in paragraph 65, the very essence of all 
rhythm. The term 'brief rise' to the climax and 'long fall' from the 
climax may best be described as 'active beginning' and 'restful end', 
respectively. Incises, being the least important of all divisions, are 
accorded active rises to the climax and a slight decrease in volume 
and speed in the fall from the climax. Members, on the other hand, 
being more important divisions, are accorded active rises but "more 
pronounced decrease in volume and speed in the fall from the climax. 


UHITT OF PHRASES 11 133-134 

The phrase, however, rises consistently from the first note to the 
primary accent (in this example on Group 17) and falls consistently 
to rest from it. In this sense the first incise in the example given 
above can be considered as a rise division preparing for the rise in 
the second incise but without losing its own fall to rest which now 
becomes a secondary fall of support. The member is brought to a 
temporary rest of support on Groups 8, 9 and 10. The third incise 
can be considered as a fall division in which the melody gains mo- 
mentum, as it were, for the grand rise of the phrase on Group 17. 
Henceforth all the groups fall to rest but rise groups (Groups 21 and 
22) are inserted to break the monotony of the fall (in the eight suc- 
ceeding groups). Too much emphasis cannot be laid upon the impor- 
tance of the active beginnings of every division. Just as the life of 
an elementary rhythm lies in its active rise note, so the life of every 
division lies in its beginning or rise to the climax. While the ends 
of the divisions are often rendered slowly and softly, the importance 
of the beginning is overlooked. The active beginning is as important, 
if not more important, than the restful end. Each division, as a 
division, has a right to its own active rise and restful fall even though 
it loses a part of its individuality in favor of the whole phrase. If 
breath must be taken in the course of the phrase, the interruption 
must be so made as not to imply an end (see par. 12). Finally, 
the last two or even three groups of every phrase are sung very 
slowly and gradually tapered off into silence. (See Fig. 92 on the 

133. Group 12 is a fall group that contains an accented syllable. 
The rise note on the accented syllable om- of omnium is more than 
an ordinary rise (see par. 107). The special treatment given to this 
note is called undulation, derived from the Latin word unda meaning 
'a wave'. It is rendered particularly brief; in fact, it is the briefest of 
all the notes in the phrase (par. 44). It is brief as a rhythmic rise and 
brief because it is sung to an accent syllable of the word. The undu- 
lation occurs only in a fall group which is followed by another fall. 
This is reasonable for if a rise group were to follow, this note could 
not be lifted and the rise group could not be respected, for nothing 
can rise twice without an intermediate fall. Undulating rhythm is 
further illustrated in the following paragraph. 

134. The grouping of a phrase in a syllabic chant can be studied 
in the Dies irae, from the point of view of the climax, of the nature 
of groups, and of the accents. (See Fig. 93 on the insert.) 

(a) In regard, to the climax: The melody in the first and third 
members is very sober, giving hardly any indication of its highest 
point. In these cases the accent of the last word is chosen as the 
climax. The second member contains a neume on an accented syl- 
lable which is undoubtedly the climax of this member as well as of 
the whole phrase. 

(b) In regard to the nature of the groups: There are ten fall 
groups in the total of fourteen groups. Such a condition would render 
the selection very calm and almost lifeless, were it not for the fact 



that the rise note of most of these groups is set to an accented syl- 
lable. This fact gives each rise definite life (par. 106). The accents 
of Groups 2 T 3, 4, 5, 10 and 11 are undulated over rise notes (par. 
133). The reason that these groups must be rendered as falls lies 
in the fact that the ictus of each is on the last syllable of a word 
(par. 119). 

(c) In regard to accents: All the accents of words except one are 
set to a rise note of the rhythm, thus giving the phrase, as was said, 
sufficient vitality to counteract the calmness of the many falls. 
The one accent which has the ictus and is therefore stressed, is the 
accent of saeclum which also forms the climax of its own member 
as well as of the whole phrase. 

135. Careful examination of the two analyses given above will 
reveal that the goal set at the beginning of the process of unification 
has been attained: we have marshalled all the notes of one phrase 
into one grand rhythmic movement — one rise and one fall. Each 
note, each group and each incise and member while retaining its 
individual character of a rise or fall, becomes a part of the grand 
rise and fall of the phrase. The individual groups and divisions have 
become a part in the uninterrupted flow of rhythm from the be- 
ginning to the end of the phrase. Through the agency of rhythm we 
have unified the notes into one living, flowing rhythmic movement. 

136. We have now succeeded in unifying (and by unifying also 
vivifying) the notes of a complete phrase. The whole process may 
be illustrated in the scheme in Fig. 94 on the insert. 


1. The process of unification must be continued until all the 
notes of one phrase are united into one rhythmic movement (par. 


2. Each note, each group of notes, each incise and member will 
retain its characteristic quality of a rise or fall entity although this 
quality is somewhat sacrificed in deference to the whole phrase 
(par. 123). 

3. Each division retains its own rise to a climax and fall from it 
(par. 124). 

4. A practical example of unifying is given in paragraphs 124- 

5. When an accent syllable occurs on a rise note of a fall group, 
which group is followed bv another fall group, that note is undulated 
(par. 133). 

6. Syllabic chants are treated in the same way as neumatic 
chants, as is to be observed in the example given in paragraph 134. 



3 ■-'■■ ^ 1 , lUu.Ttl t 

W» "V 

a-nimas omni- um fi-de-li-um de-fun-ct6 rum 

• • 

5 1 1 

I 1 i 

i 1 

6 i i 


1 . 

i 1* 7 2 3 

i i 

6 1 

a- ni- 

mas omni - 

um fi 










16 17 



Fig. 91 

665 65 11 61 55 4. 

20 21 22 23 24 25 




■ ■' 



i" 1 "^ UMb SIt 

Bsol- ve, * D6-mi-ne, 

a- nimas omni- um fi-de- li- um de-functo- rum 
Fig. 92 

I-es i-rae, di-es il-la, Solvet saeclum in favil-la : Teste Da-vid cum Si-bylla. 

Fig. 93 

elementary rhythm 

simple rhythm 



members (2 incises) 

phrases ■ I 

(2 members) 


■ a 

■ ■ ■ ■ ■ 

■ r 


Fig. 94 

II 137-140 



137. The principles to be set forth in this chapter were already 
hinted at and referred to in the preceding chapter. But in order to 
enlarge upon them and to set them in good order, a new chapter is 
devoted to their exposition. This chapter, then, will treat of how 
one division is joined to another. One phrase is never joined to the 
next since the phrase indicates a melodic and, in the majority of 
cases, also a textual completeness. Incises and members are, how- 
ever, joined in a more or less intimate fashion. 

138. The most intimate of all joinings takes place when the first 
note of one division is isolated and recedes to become a member of 
the last group of the preceding division. In this case the end of one 
division and the beginning of the next are fused into one group. 
There is an example of this most intimate joining in the first two 
incises of the Requiem Tract which was discussed fully in the last 
chapter. Group 5 of Figures 85 and 87 is a time group of three notes, 
the first two of which are the dotted note and the end of the first 
incise and the third of which is the isolated sol, the beginning of 
the second incise. This kind of joining can of course be made only 
when the second division begins with an isolated note. If it begins 
with an ictus note, that note is the first of a new group which in 
turn will be either a rise or a fall group. 

139. When the second division begins with an ictus note, that 
note is the first of a group. The next most intimate joining would 
take place if that group could be rendered as a fall group continuing 
the fall of the preceding division. The third and fourth incises of 
the Requiem Tract can be joined in this way.- There is no reason 
why Group 16 in Figure 92 cannot be rendered as a fall. The melody 
here does not lie higher than that at the end of the previous incise. 
Objection might perhaps be raised that the ictus note is on a sec- 
ondary accent (de- of defunctorum). But that need not change the 
interpretation for one who prefers to render the group as a fall. 
For this reason it was suggested (par. 131) that Group 16 be consid- 
ered a fall group — such a treatment would bind. the two incises 

140. The third and least intimate of all joinings takes place 
when the second incise or member begins with an ictus note which 
is the first of a group that must be rendered as a rise. In Fig. 92 
this happens in the third incise or, in other words, at the beginning 
of the second member. Group 11 is a rising melody on an accented 



syllable; hence it cannot be considered in any way except as a rise 
group. It is quite evident that this kind of joining is the least inti- 
mate, for the last group of one division must of necessity be a fall 
group while the first of the next division is a rise group on an accent. 
There is such a difference in the nature of the two that in rendition 
there must be no doubt as to where one division ends and the other 
begins. Group 11, beginning the second member, is like the return 
swing of a pendulum, whereas the other joinings are all within the 
same swing of the pendulum. 

141. In paragraph 137 it was stated: "One phrase is never joined 
to the next since the phrase indicates a melodic and, in the majority 
of cases, also a textual completeness." The phrase always indicates a 
melodic but it may not always indicate a textual completeness. The 
first phrase of the Requiem Gradual is complete in regard to its 
melody and text. The words of the first phrase, Requiem aeternam 
dona eis, Domine, express a complete idea and the melody ends on 
the final of the mode. But the second phrase consists of the words, 
et lux perpetua, which do not express a complete idea. The idea is 
only completed when the words of the third phrase, luceat eis, are 
added. Hence, if the text is preferred to the melody, there ought to 
be no interruption at the phrase mark between perpetua and luceat. 
But if the melody is preferred to the text, an interruption, that is, 
a retard and a gradual descend into silence with a pause of two 
beats before the next phrase, is to be made. The director should use 
his discretion in such rare cases. 


1. Although all divisions of a phrase are unified into one phrase, 
one division may be joined more or less intimately to the next 
(par. 137). 

2. The most intimate of all joinings takes place when a division 
begins with an isolated note which becomes a member of the last 
group of the preceding division (par. 138). 

3. The next intimate joining consists in rendering the first 
group of the next division as a fall group, continuing the fall at the 
end of the preceding division (par. 139). 

4. The least intimate of all joinings takes place when the first 
group of a new division must be rendered as a rise group (par. 140). 

5. In rare instances where the text is not complete at the end of 
a phrase, the text or the melody may be given preference according 
to the discretion of the director (par. 141). 


1f 142-143 




142. The principles of rhythm of the foregoing chapters can be 
applied to special figures where they can reveal delicate and beau- 
tiful nuances which might perhaps be lost if attention were not 
directed to them. 

143. In passages where syllables are adorned with single neumes, 
it is at times difficult to impart a definite rhythmic flow. A hint of 
sluggishness might'characterize the flow in passing from one syllable 
to the next. Whereas the first note of the neume is generally the 
fall note, the second (or third, in a group of three) is a rise note and 
the rises are particularly brief. Therefore, any tarrying on last notes 
of neumes would have the effect of prolonging the rise notes, robbing 
the notes themselves of their liveliness and brevity and in turn de- 
stroying the very nature of Gregorian rhythm. The beginning of 
the Requiem Offertory versicle offers an example. For illustration 
and clarity, the length of the notes can be exaggerated in order to 
insure an unmistakable rise and fall movement; that is, the rise 
notes should be sung quite briefly and the fall notes quite long. This 
section has been transcribed into modern notation (Fig. 95) to show 
more clearly the variety of length . 


1 2 




pre - ces 
5 6 


D6-mi - ne, lau - dis of- fe - ri - ] 

8 9 10 11 12 13 14 

of- fe - ri - mus 
13 14 15 

Fig. 95 

All the fall notes are given the value of a quarter note, the rises are 
represented by eighth notes. Rendered in this way, the rise of the 
rhythm on the brief notes and its fall on the long notes are clearly 




felt. All word accents are sung to the rise notes of Group 1, 6, 7 and 
13. These notes are particularly brief becajuse of the accent. The 
ictus of Groups 5 and 11 are stressed. This passage (or any passages 
that might lack rhythm) should be sung as written over and over 
again until the rhythm is definitely felt; the notes should be given 
their time value as called for in the chant notation and the flow 
will be insured. 

144. For practice, in order that the student feel a true rhythmic 
flow, it was recommended that rise notes be abbreviated and fall 
notes prolonged. Frequently, however, this very device is indicated 
by the notes themselves: the fall note is lengthened by an episema. 
by repetition or by length in the neumes. In the Requiem Offertory 
the following are found: 

(in-) fer- (ni) (le- o- 

(nis)(ob-) scu- (rum) 


e - 


(as) (se - mi -) ni 'Ho-sti-) as 

Fig. 96 

On the syllable -i of perfrui at the very end of the Requiem Tract 
we find a four-fold repetition of the rhythmic figure: 

ii * * a * * 

1 J'M 1 ■ ■ ■ " { ♦ — ?i 1 J ^ » . — 

di- ne * pcr-fru- i. 

Fig. 97 

In the above illustration it will be noted that the falls of each rhyth- 
mic figure are definitely prolonged in order to guarantee the rhyth- 
mic flow. Notice how the flow is insured on the syllable -tu- of 
beatitudine in the Requiem Tract: 




■ ■ ■ 

be- a- ti- tu- 

Fig. 98 

The first and second la are long to lift the sol that occurs between 
them. In the Dies irae (third stanza) there is a melodic inversion of 
five notes; that is, the rhythmic fall is five notes higher than the 
rhythmic rise: 


%_ 3- 


Li-ber scriptus pro- fe-re- 

Fig. 99 

In spite of the melodic inversion, the rhythm is guaranteed if the 
fall note is lengthened as indicated and the rise note rendered briefly 
as usual. 

145. The episema (pars. 16, 50) indicates that the note (or notes) 
over which it stands is to be prolonged slightly beyond its ordinary 
length. A fall note is long by nature; the episema will extend its 
length to a slight degree. A rise note is brief by nature; the episema 
will prolong it slightly. In other words, the episema does not render 
all notes over which it appears equally long: it merely extends their 
natural length. Furthermore, the groups that appear at the be- 
ginning of incises and members are quite active and brief; those at 
the end are by nature restful and long. The episema will give slight 
prolongation to each. The Kyrie of Mass I furnishes an example. In 
the Christe of that composition there is a torculus with episema on 
the accent of the word. This torculus is quite brief by nature since 
it occurs at the beginning of the division and on an accent of the 
word. The episema prolongs the torculus somewhat. But there is 
also a torculus with episema on the last syllable of the second Kyrie. 
This torculus is naturally long because it occurs on a last syllable 
and near the end of the division. Again the episema prolongs its 
value somewhat. But there is a vast and pronounced difference in 
time value between the two: the first is quite active even with the 
episema, the other very restful because of it. 


* 146-147 






i -M 

ft^r i lv 

y- ri- e * e- le- i-son. iij. Chri- ste 






' t » , * ' .. 


le- i-son. it/. Ky- ri- e e- le- i- son. ij. Ky- ri- e 




le- i-son. 

Fig. 100 

146. The rise notes of the chant are rhythmically brief but are 
not all of the same brevity. The third note of a group of three is 
slightly briefer than the second because it is a more definite rise than 
the second. Furthermore when an accent occurs on that third note, 
the lift of the accent renders it even briefer (see third note of rise 
in Groups 5. 9. 10. 13. of Fig, 93). There are nine instances in the 
Requiem Offertory, including the three already mentioned, in 
which the accent of a word 'either primary or secondary) occurs on 
the third of a group. 1 (See Fig. 79.) 

147. The chant delights in 'leaps' on and off a syllable. The leap 
is brought about by setting the notes on a syllable in such a way 
that the first is brief, the second long and the third brief. The first 
represents the spring, as it were, upon the syllable: the second. the 
momentary settling upon it; the third, the leaping off to the next 
syllable. These leaps may occur in an ascending or descending 
melody. The salicus (leap) is generally used for the upward leap. 
The two syllables ae- of aeternam and do- of dona of the Requiem 
Introit are so treated. The In- of luceat and the non of the Gradual 
are similarly treated. The Do- of Domine of the Tract is leaped upon 
but here the effect is produced by the pressus. Sometimes it is the 
pressus that makes the downward leap, as with -ter- of aeternam in 
the Introit. The -ter- of the aeterna in the Gradual versicle is leaped 
upon, as shown by the ictus on its second note. But such a settling 
on the syllable is not as long as when represented by the pressus. 
Of course, when the leap is on an accented syllable, the settling is 
stressed. In the Introit the ae- of aeternam and the do- of dona are 

^he words are: Christe, libera (first . Abrahae, promisisti (secondary accent^, 
tibi, Domine second 1 , ojferimus. eas last . Domine (last . 



K 148-149 

adorned with the salicus. The ictus of the first is long, of the second, 
both long and stressed because the first is on a weak syllable of the 
word, whereas the second is on the accented syllable. 1 

148. The first note of length takes the ictus (par. 84) except 
when the second is marked as ictus. Although the latter occurrence 
is relatively infrequent, it does appear in the Requiem Gradual and 
in other Graduals similar to it. In the Requiem Gradual on the last 
syllable of the last word, timebit, there is found this rhythmic figure: 

C . P LJ l i ± 


fl s -?♦'• JO 


time- bit. 

Fig. 101 
Transcribed into modern notation, it appears as follows: 


Fig. 102 

In order to insure the correct rendition of this nuance, we must pro- 
long the notes that are marked long (designated by an asterisk) 
and render the others quite briefly. But we shall observe that two 
do's are adjacent, the first being brief and the other long. The only 
way to distinguish one from the other is to alight briefly but deli- 
cately upon the first and, by a slight repercussion (a delicate inter- 
ruption of breath), to sustain the second. 

149. In paragraph 120 we learned that one group of notes grows 
out of the other. This is especially true where there is little variation 
in melody. In other words, in passages where the melody contributes 
little or nothing to the rhythmic flow, other features of dynamics 
and agogics must be called upon to help bring it about. Each group 
in these sections is exaggerated in its rise or fall qualities. So, for 
instance, in the third incise of the Tract there is very little variety 
of melody (Fig. 89). But if each group is given definite dynamic qual- 
ities, the rhythm must become evident. Thus, a beautiful melody 

'The student may examine these compositions for himself in the figures 
which have been given in the preceding chapters. 




like that on the ju- of Justus in the Gradual would become stagnant 
unless the dynamics were emphasized. That melody is transcribed 
below with the rhvthm indicated. 


♦^ A^fl 

m — m 

•— •- 



3 1 

3 2 

1 1 6 

1 1 6 

1 . 

1 1 

1 6 

1 2 

32 1 

3 2 1 

2 6 



























Fig. 103 

In this short passage, groups are repeated. Groups 3 and 4 are alike 
in rhythm and melody; Groups 5 and 6 differ only in notation (one 
a dotted note, the other a repeated note) ; Groups 9 and 10 are alike 
in melody. But rhythmically, that is, dynamically, they differ: 
Group 9 is louder and faster than Group 10. Group 4 is louder and 
more active than Group 3 because the former is nearer the climax; 
Group 10 is softer and more restful than Group 9 because the latter 
is the climax and the former prepares for the fall. The difference in 
dynamics is indicated in some degree by the vertical dotted line 
that appears within the dynamic angle. Unless the differences in 
rhythm and melody are made evident, beautiful passages such as the 
one illustrated here will be deprived of their inherent beauty. 
Without rhythm they can have no life or unity. 

150. In paragraph 119 it was stated that ascending melodies 
are to be rendered as rise groups and descending melodies as fall 
groups. Moreover, rise groups are active because they are rendered 
briefly and fall groups are restful because they are rendered com- 
paratively longer. This introduces the feature of spirituality into 
the chant. This concept is often difficult to understand for we are too 
accustomed to think of heavy things rising slowly and falling fast 
and light things, on the contrary, rising fast but falling slowly. The 
ascending melodies are rendered as active rise groups and the de- 
scending melodies as reposeful fall groups and as such never to be 



151. As the dynamic angle opens and closes very gradually, so 
do the dynamics of the chant increase and decrease gradually. The 
transition in volume from one note to another or from one group to 
another, is never abrupt or sharply defined. It is a bad habit among 
singers to render the higher notes louder than the lower. In the 
chant this habit is intolerable. In the Requiem Gradual the syllable 
-is in the phrase luceat eis finds its climax on high fa but the rise to 
it is very gradual. So also the transition from the second to the last 
incise on the syllable -ter- of aeterna (from la to high re) must be 
performed in a very delicate fashion. The la forms a fall group which 
is indeed quite soft but the next high re should not be much louder. 
In chant nothing must divert the perfect calm from the flow of the 
melodies. They reflect the "peace that the world cannot give", that 
is sought and found in the House of God. 

152. It has been maintained throughout this book that the be- 
ginnings of divisions are active and the ends are reposeful. Activity 
without repose implies haste but, contrariwise, repose without ac- 
tivity reflects sloth. A temperate mixture of both must be introduced 
in every division. There seems to be no difficulty in rendering the 
end of a division reposefully. But the beginning of the next division 
must be imbued with new life. Care should be taken, therefore, to 
begin each new division actively. Where this is not done, the tempo 
decreases gradually to such a degree that the phrase loses its life 
long before it should descend to rest. In the Requiem Tract, there- 
fore, the word Absolve is ended rather peacefully but the next word 
Domine is a new beginning which must reflect activity. Likewise 
the new member animas omnium fidelium begins in a lively manner 
and falls to a temporary respite on its last syllable, but the beginning 
of the next word defunctorum is full of life since it forms the climax 
of the whole phrase. 




1. Delicate nuances more thoroughly explained (par. 142). 

2. Syllabic chants are enhanced by careful rendition of rise notes 
(par. 143). 

3. Lengthening fall notes and abbreviating rise notes often brings 
out the rhythm (par. 144). 

4. The episema. slightly lengthens the note or notes which it 
effects (par. 145). 

5. Rise notes are not all of the same brevity (par. 146). 

6. The chant delights in leaps (par. 147). 

7. When the second note of length has an ictus, a slight repercus- 
sion is necessary in its rendition (par. 148). 

8. Dynamics are emploved to procure rhythm when the melody 
fails (par. 149). 

9. When rising groups are rendered actively and fall groups repose- 
fully, the chant is invested with a definite spiritual character 
(par. 150). 

10. Dynamics in the chant are always smooth (par. 151). 

11. Active rises and restful falls cloth the chant with life without 
rush, and rest without sloth (par. 152). 


H 153-156 


153. The directing of the rhythm of the chant by hand is called 
chironomy, a word derived from the Greek words cheir, meaning 
'hand' and nomos, meaning a 'usage' or 'direction'. 

154. Chironomy depicts the rhythmic flow of the individual 
notes and of note groups, and also indicates the volume levels at 
which the group is to be sung. It delineates the individual notes in 
so far as the hand is always in a relatively lower position on the 
first note of each group than on the second (or third in a group of 
three). The first note of the group is the support note from which the 
second (or third in a group of three) rises. The group is represented 
by curves made upward from right to left (for a rise group) or down- 
ward from left to right direction (for a fall group). The dynamics 
are represented by the height at which that group is described. 

155. The simplest form of rhythm consists in an alternation of 
rise and fall notes. An example of this form is the simple Dies irae 
where the rises occur on the accents and the falls on the final 
syllables. The hand is raised and a curve described from left to right 
like an inverted U over the accented syllables and lowered on the 
ictus notes. This is more familiarly known as undulating rhythm 
(par. 133). The rise notes are on the accents but in fall groups. The 
hand movement finds its origin in the nature of the syllables as found 
in the form of accent signs. This undulating movement has already 
appeared in joining the accent marks of the words as in Fig. 46. In 
describing this form of rhythm the hand is lifted lightly but briskly 
in a curved motion over the accent syllables or rise notes and is 
allowed to drop of its own weight on the last syllables. The re- 
straint implied in the words 'drop of its own weight' suggests that 
the ictus on weak syllables is never stressed. 

156. The next simplest form of chironomy consists in an alter- 
nation of rise and fall groups. This form is quite rare since the melody 
is hardly ever content to move in so regular a fashion. It is found, 
however, in the Adoro Te. In this kind of rhythm the hand is raised 
in a curved upward line from right to left for the rise groups and in a 
downward curved line from left to right for the fall groups. A rise 
group of three is delineated in a larger curve, the hand being highest 
at the count of three; a fall group of three is described in a larger 
downward curve, the hand, in this case, too, being highest at the 
count of three. It is important to note that in every group the hand 
is lower at the count of one than at the count of two (or three in a 
group of three). 



157. The most common of all rhythmic formations consists in 
an indiscriminate succession of rise and fall groups. It is to be noted 
that when one rise group is followed by another, the junction of the 
two groups will be delineated in a small circle described upward from 
the left and downward to the right. The upward movement de- 
scribes the one or two rise notes of the preceding group while the 
downward movement depicts the ictus of the second group. 

158. When one fall group follows another, the hand is raised 
in a rather horizontal curve describing the one or two rise notes of 
the first group and allowed to drop gently to depict the ictus of 
the next group. 

159. It is to be understood, of course, that the hand does not 
progress in space as notes do on paper. Very little space is needed to 
delineate the figures as represented in the last illustration. In any 
case, the hand need never extend beyond the dimensions of the body. 
The chironomy in the monastery at Solesmes is so unobtrusive that 
no one is aware of the choirmaster's direction since he stands with 
his back to the congregation. It is said that he uses nothing but 
the wrist and index finger for even the most emphatic kind of 
direction. Such a procedure, however, presupposes great skill on 
the part of not only the director but the singers as well. 

160. The director should not lay too much emphasis on grace of 
the chironomy. At the final performance it is the director's duty to 
start his singers together, to keep them together throughout, and 
to bring them to the finish together. Any other instructions should 
be given them prior to their appearance and should be practiced 
during the rehearsals. Furthermore, some directors execute the 
most graceful chironomy while the minds of the choir members 
wander and their eyes are fixed on some foreign object. Other direc- 
tors need only to chironomize in a calm and unobtrusive fashion: 
the rendition is good because they have the attention of their choir. 
Whatever signs of direction are used, they are good if they are under- 
stood by the choir. An attentive, well-schooled choir needs very 
few of them. 


1. A rise group is delineated in an upward curve beginning at 
the right and bearing towards the left; a fall group is delineated in 
a downward curve beginning at the left and bearing towards the 
right (par. 154). 

2. The undulation is described by a brisk curve made in an up- 
ward and downward motion over the accent to be lifted (par. 155). 

3. In delineating a group the hand is always lower at the count 
of one (the ictus note) than at the count of two ; in describing a group 
of three the hand is raised higher on the count of three than on the 
count of two (par. 156). This is true of both rise and fall groups. 


If 161-164 




161. In applying the foregoing principles of Gregorian rhythm 
to modern music, it should be noted that the modern system of 
notation, while practical in itself (especially since the introduction 
of polyphonic music) does, however, detract from the true nature 
of rhythm. 

162. Every rhythm in modern music is severed into measures by 
the bar line. No measure is a natural unit or complete entity in 
itself- The last note of every measure is the first note of a rhythm 
and the first note of every measure is the last note of a rhythm. This 
has been brought about by setting the bar line before the original 
ictus note. But the ictus note is really the fall or end of the rhythm. 
The bar line or the divisions of the notes into measures is merely a 
convenience for setting the rhythmic beats in strict mathematical 

163. In modern music the first note of every measure is the 
'accented' beat and as such represents the fall of the rhythm. It 
necessarily follows that the last note of every measure is the light 
or 'unaccented' beat and as such represents the rise of every rhythm. 
But it has already been demonstrated (par. 68) that the fall or 
'accented' beat is in reality the end of every rhythmic movement 
and the 'unaccented' beat is the beginning of every rhythmic move- 
ment. In modern music, however, rise and fall notes, or the 'un- 
accented' and 'accented' notes are unnaturally placed at the end 
and at the beginning of each measure respectively. Thus the bar 
line divides, or rather breaks up, every simple rhythm. 

164. An example of this apparently confusing (but in modern 
music, practical) method, will clarify the exposition. In paragraph 
71 the simple phrase Deus Pater Jesu Christi was used to show the 
most natural place for the rhythmic ictus (i.e. on the last syllable of 
the words) : 

5 f> b § Peej? 

* ■ ■ ■ * m * 

De - us Pa 

ter Je - su Chri - sti 
Fig. 104 


r :t< 


I: :he 

set in modern measures. the bar lines : be 







•r Te 

so Chn- 

Fiz. 10: 

tvery w 

t " .era te 
Hence :! 

_ . 

i >• 

red as the 


• - ■ s 

in setting 


urs-c. .r.c .Ti.Ccrr.; rcu.u never 

rented", the stressed beat. It is 
ion to stress the last syllable of Latin winds 
ce for the rhythmic beat was on the accer. be 1 
rhere it is invariably found today. But in 
1 is not necessarily heavy or accented' but 
therefore, no contradiction nor distortion 
he list svllables in the chant. 

It-. Tne v.-:rc a:: 
of the measure This rente: 

-. the rail of the rhythm 
meaning- of the w:rd actenu 
ous to its more modern meat 
syllable of a word is often an 
other syllable to the beginni 
: ears at relatit a t : tae rhy 
; -. i : : alir Lu'.'.cb\ : f Brah 
s:ra:-r = htw wrrris and me a 

s are generally set to the accented beats 
lets every accent heavy since it concurs 
m This may account for tne change in 
nt from its former implication of mtlodi- 
aning of stressed. In modern rendition one 
assigned to the end of a measure and the 
wing f the next. Thus again the bar line 
rule Bow. Examination of the simple 

with the usual English text demon- 

es are d:sl::ated: " 





Lui - 1; 

bv and good - nieht, with 

ros- es 









es be 

decked is 



f \ h S 





ba - bv's wee 

bed ; Lav thee down now and 

Fig, 106 








P ''J> J' i 

rest, may thy slum - ber be blest, lay thee 







down now and rest, may thy slumber be blest, 

Fig. 106 (cont.) 

166. Modern music uses the whole note as its unit of measure. 
Every other note is a fraction of this note. 

O _ whole note 
O _ hal 

ilf note 

J _ 

quarter note 

eighth note 

Fig. 107 

In vocal music certainly it would be more practical and rational to 
consider a shorter note as a unit. A shorter note would better express 
the time value of a syllable in discourse (par. 46). If in practice the 
the eighth note were considered the unit, many rhythmic difficulties 
would soon be solved : all the rhythmic units of the larger note could 
then be broken down into their proper value and the rhythmic ele- 
ments of each could readily be recognized: 

J. J> ■ 
jti t 

Fig. 108 

Furthermore, shorter notes would add life to the movement, for 
in the longer notes the rhythm can easily be lost. If we realize that 
there are e,ight movements in a whole note, the rhythm must of 
necessity continue to flow: 


ji r> h h h jj \) J) 

= mmmmmdm-mr 

l t i t i t l t 

Fig. 109 


11 167-168 


167. Even though modern music bases its rhythmic principle on 
stress, it cannot deny the importance of duration. The last notes of 
each measure are brief (because they are rhythmic rises) and the 
first notes are long (rhythmic falls). In the Lullaby (Fig. 106), in 
only one case (from measure 2 to 3) is the rise note almost equal in 
time value to the fall (par. 44) : 

Last note 

First note 


; note 

First note 

of measure 

of measure 

of measure 

of measure 

, J> 

, J- 



3 J 

, J 

3 J 




3. J" 





Fig. 110 

168. The larger musical divisions are seldom designated. There 
are no subordinate divisions to indicate the structure; they must be 
inferred from the text and its meaning, from the music and its or- 
ganization. It can be said that in general every sentence of text 
with its setting corresponds to a musical phrase in the chant. It 
can be broken up into subordinate divisions according to the struc- 
ture of either the sentence or the musical phrase. Each of these 
subdivisions must possess its own rise and fall, subordinated to the 
sentence according to the principles laid down in Chapter X. The 
Lullabv can then be divided and subdivided as follows: 

1st member 1 


1st phrase*! 


I 2nd member', ul 

3rd member^ g^ 

incise: Lullaby and good-night 
incise: with roses bedight 
incise: With lilies bedecked 
incise: is baby's wee bed. 
incise : Lay down now and rest 
incise: Mav thy slumber be blest 

2nd phrase /7th and 8th 

4th member j incises . rep€ tition of 5th and 6th 

The dynamics can be represented as follows: 

Lul-la-by and good-night, with roses bedight, 
Fig. Ill 



with lil-ies be-decked, is baby's wee bed. 

Lay thee down now and rest, May thy slumber be blest. 

Lay thee down now and rest, May thy slumber be blest. 

Fig. Ill (cont.) 

2 3 3 

169. In the simple rhythms of — , — and — time there 

is one ictus to each measure, occurring on the first beat. In the com- 

4 6 
pound rhythms, however, such as — , — , etc., there are two ictus 

4 o 

notes, a primary and a secondary, in each measure. The rhythms in 

— , — and (as some say) — occur in groups of two ; the others, 

in groups of three. Every group in modern music, as in chant, should 
be rendered either as a rise or fall group, according to the disposition 
of the melody and the nature of the text. In this way every group 
before the melodic climax aids the flow to it; and every group after 
it, the flow to rest from it. 

170. Very often conductors will rightly disregard the bar line 
in the performance of modern music. By so doing they can give 
greater attention to the musical phrase and its subordinate divi- 
sions. In this way a much truer and clearer picture of the rhythmic 
flow can be drawn. 

171. The Gregorian system of rhythm is not incompatible with 
modern music if its principles are applied correctly. In point of 
fact, their application will greatly enhance the beauty of modern 
music. Moreover, in comparing the chant and modern music, the 
true nature of rhythm can be more clearly explained, more easily 
understood and more readily accepted by modern musicians. 



1F 172-174 




172. Gregorian melodies are written in definite tonalities called 
modes. A tonality is a musical effect produced by the preference 
given to the various notes of the scale. It happens that some series 
of notes reflect joy, others, sorrow; some, gallantry, others, meek- 
ness. The choice of notes and the preference given them in a musical 
selection is called a tonality. One tonality differs from the other in 
that the half-tones are in different relation to the tonic and dominant 
of the mode. In modern music two scales are used, the major and 
the minor. The major has for its tonic do, for its dominant sol; its 
range extends from do to do and its half-tones occur from the 3rd 
to the 4th and from the 7th to the 8th degrees. The minor has la 
for its tonic, mi for its dominant; it extends from la to la and its 
half-tones occur from the 2nd to the 3rd and from the 5th to the 6th 
degrees of the scale. Each scale, therefore, encompasses an octave. 
The major has do for its tonic and the fifth above for its dominant; 
the minor has la for its tonic and the fifth above for its dominant. 
The former, we may say, is built on do, the latter on la. 

173. Two Gregorian modes are built on each of the seven notes 
of the scale: do, re, mi, fa, sol, la and ti. This brings the number of 
modes to fourteen. By changing ti to te (ti flat), however, Mode IX 
can easily be transposed to Mode I; Mode X to Mode II, etc., as 
will be explained in paragraph 178. Consequently it is generally said 
that there are actually only eight modes in the chant. In the dis- 
cussions which follow, numbers will be used instead of the conven- 
tional sol-fa names to simplify and clarify the exposition. 

174. Every mode has three elements: a final, a dominant and a 
range. The final is the note on which melodies of that mode end. A 
melody always falls to rest on the final of the mode. There are 
(assuming the number of modes to be eight) four finals: 2, 3, 4, 5: 
2 is the final of Modes I and II; 3 of Modes III and IV; 4 of Modes 
V and VI; 5 of Modes VII and VIII. 1 The dominant of the mode is 
the note around which the melody flows. It is, as its name implies, 
the 'dominating' note of the mode. In the odd-numbered modes 
(I, III, V and VII) the dominant is always the fifth above the final; 
in the even-numbered modes (II, IV, VI and VIII) it is the third 
above the final. The range is the octave within which the mode is 

^he finals are set in bold face type; the dominants are in parentheses. 


MODES . H 175-176 

written. A melody may not always remain strictly within its range; 
it may exceed the range either above or below the octave. Gregorian 
melodies, however, remain for the most part within the octave. 

175. Modes are divided into two classes: the authentic and the 
plagal. The authentic modes are the original; the plagal are the 
modes derived from them. The numbers set before the staff desig- 
nate the number of the mode in which the melody is written. This 
system is used in the Liber Usualis, the Graduale Romanum and 
other church music books. Intr. 6 signifies that the Introit is in 
Mode VI; Offert. 2 signifies that the Offertory is in Mode II, etc. 
The odd-numbered modes are authentic; the even-numbered, plagal. 
The relationship of the two modes consists in this: both have the 
same final and both are built on the five notes ascending from the 
final. The authentic completes its range by extending the notes up- 
ward to the octave above and the plagal, by extending downward to 
the octave below. For example, the final of Modes I and II is re. 
The five notes built on the final are common to both modes, the 
authentic and the plagal. Five notes built on re form the common 
or central fifth of Modes I and II: 

2 3 4 5 (6) 
I I 

The authentic mode (in this specific case, Mode I) is continued up- 
ward to the octave of re: 

2 3 4 5 (6) 7 1 2 
I _J 

and the plagal (in this specific case, Mode II )is continued downward 
to the octave of la: 

1 2 3 (4) 

The relation of the two modes and the importance of the central 
fifth may be seen when both are joined: 

6 7 1 2 f (4) 5 (6) f 1 2 
I I 

176. Since the dominant of the authentic modes is five notes 
removed from the final, the melodies of these modes are quite active 
because of the large space in which they move. On the other hand, 
the dominant of the plagal modes is only a third from the final; 
the space is so limited between the final and dominant that plagal 
melodies, are naturally expected to be quite sober and quiet and 
generally more restful. The authentic modes often delight in inter- 
vals of fifths; that is, from the final to the dominant, as re - la or 
la - re in Mode I and sol - re or re - sol in Mode VII. 


1 177-178 


177. The system of the modes as built according to the principles 
laid down in paragraph 175 is as follows: 

Central Fifth 















































178. It was found that by a slight change of ti to te, in Mode I, 
Mode IX could be changed to Mode I; Mode X to Mode II, etc. 
Thus the number of modes is reduced from fourteen to eight. The 
only difference between these two modes is that Mode I has the 
half-tone between its sixth and seventh degrees, and Mode IX 
has the half-tone between its fifth and sixth degrees: 

Mode I: 









Mode IX: 










But by flatting the ti of Mode I the succession of intervals becomes 
exactly that of Mode IX: 

Mode I: 
Mode IX: 

Similarly Modes X, XI, XII, XIII, XIV become Modes II, III, 
IV, V and VI respectively. It is apparent then that the half-step 
in each mode in relation to its final and dominant gives a different 


MODES 1 179-180 

Authentic Modes 

I 2 3^^4 5 (6) 7 1 2 

III £~^\ 5 6 (7) 1 2 3 

V 4 5 6 7 (i) 2 3 4 

VII 5 6 7 i (2) 3 4 5 

Plagal M odes 

II 6 J \ 2 3^ (4) 5 6 

IV 7 1 2 3 4 (5) 6 7 
VI 1 2 3 4 5 (6) 7 i 

VIII 2 3*"^4 5 6 (7) i 2 

179. During the course of the centuries when the ear became 
trained to regard ti as the 'leading tone' because of its great affinity 
to do the dominant of Modes III and VIII, instead of remaining 
H, was changed to do. The ti is now known as the ancient dominant; 
the do as the modern dominant of these modes. Furthermore, it is 
said also that when the dominant of the authentic mode was raised, 
the dominant of the plagal modes changed accordingly, so that the 
ancient dominant of Mode IV is sol and the modern dominant, la. 
Moreover, some melodies contain such a mixture of dominants that 
it is at times difficult to determine which is the dominant; often 
there are disputes among the authorities as to the correct mode of 
the melody. This, however, is of no great concern since the chief 
duty of the singer of Gregorian Chant is to sing the notes as written, 
regardless of the modes to which they are assigned. 

180. Many people believe that every mode reflects a certain 
'mood' or characteristic according to the sentiments it arouses in 
the minds of the listeners. Adam o^Fulda (c. 1450 - c. 1537) says: 

"For every mood the first will be good; \ 

The second so tender to grief; 

If anger the third one provoke 

Then the fourth will bring the relief. 

The fifth will be the mood for the joyous; 

The sixth one the pious will prize ; 

The seventh is pleasing to the youth ; 

But the last is the mood for the wise." 
Johner 1 maintains that Mode I expresses a "happy admixture of 
strength and firmness"; Mode II "forms melodies that are exceed- 
ingly agreeable"; Mode III is "characterized as fiery and stormy 
because of its wide intervals and leaps"; Mode IV is "persuasive 
and best fitted to express fervent supplication"; Mode V is "almost 
inexhaustible in its power of producing ever new and charming, 
and often enchanting, transformations of the few fundamental 
forms"; Mode VI is one "that moves to tears but tears rather of 
sweet j oy than of sadness"; Mode VII is "aspiring and cheerful 

1 Johner, Dominic Dom. A New School of Gregorian Chant. Ratisbon, Fred- 
erick Pustet, 1925. 



and with its quick passages is a true figure of the restlessness of 
youth" ; and, finally, Mode VIII "presents in its melody a movement 
that is calm and stately." 

181. No mode, and no melody, can be said to be unreservedly 
characteristic of a particular mood, for the same melody is some- 
times used Tor occasions of quite opposite sentiment. The joyful 
Regina coeli, the Easter antiphon to the Blessed Mother, and the 
solemn Requiem Introit are both in Mode VI. Furthermore, the 
melody of the Gradual for the Requiem Mass (with slight variations) 
is used also in the four Graduals of the Saturday in Ember Week of 
Advent, in the Mass for the Vigil of Christmas, in the Mass on 
Christmas (Midnight Mass), in the Haec dies Graduals of Easter 
Sunday and Easter Week, in the Mass of a Confessor not Bishop, 
in the Mass for the Saturdays in honor of the Blessed Virgin Mary, 
in the Masses for the feasts of St. Thomas and St. Barnabas, Apos- 
tles, of St. Joachim and St. Jerome, for the XXI Sunday after 
Pentecost, and also for the Nuptial Mass. It would be difficult to 
find a characteristic common to all these occasions, which warrants 
the use of the same melody. 

182. In the study of the modes, since the dominant is the note 
around which the melody flows, it is profitable to learn the tones of 
the sol-fa scale around the dominant. For example, in Mode I whose 
dominant is la, the notes adjoining the dominant are sol and ti: 
sol la ti. Singers may find some difficulty in singing sol la ti but 
the pattern do re mi is exactly like it, consisting of two whole tones. 
The tone sol is called do; the series do re mi learned in all of its 
variations: do re mi do, do mi re do, etc. When it is translated to 
the original sol la ti, the la should be emphasized since it is the im- 
portant note. The high do is next added, forming the series sol la 
ti do which corresponds to do re mi fa. The interval sol - ti might 
cause difficulty but when translated into its parallel do - mi, it can 
be sung easily. Next, high re is added : la ti do re. The interval re -la 
is important for it is composed of the dominant and the highest 
note of the mode. The next step would consist in the study of the 
notes around the final, re: re mi fa sol. There should be no difficulty 
here, especially since the series can begin on do: do re mi fa sol; the 
notes re mi fa sol should be emphasized. Then the two can be joined 
(re mi fa sol la ti do re) and the notes of the entire mode learned. 
Since re is the final and la the dominant, the chord re fa la is very 
important. Mode I lingers upon these notes in intervals, as re fa re, 
fa la fa, re la re, etc. For models of this mode, the Kyrie of Masses 
IV, IX and X can be examined; for the transposed mode, the Kyrie 
of Mass XL 

183. When two or more modes are learned in the above method, 
the other modes are learned very easily, for the same notes and in- 
tervals recur constantly. Little need be said here about the nature 
of the intervals. Too much stress should not be laid upon the names 
of intervals: the fact that do - mi is a major third or re - fa a minor 
third means nothing to non-musicians. Music must be practical 
in order to be understood and appreciated. Theorizing in music pro- 
duces two evil effects: it makes music unfruitful and distasteful. 



If 184-185 

184. Typical melodies of the various modes, taken from the 
Commons of the Masses, are classified below; the melodies of these 
chants seem to characterize the mode in which they are written. 

Mode I Kyrie X, Sanctus XIV, Agnus XIII 

Mode II Gloria XI, Sanctus XI 

Mode III Kyrie II, Gloria XIV 

Mode IV Kyrie III, Gloria IV, Sanctus III, Agnus V 

Mode V Kyrie VIII, Gloria VIII, Sanctus IX 

Mode VI Sanctus VIII, Agnus VIII 

Mode VII Kyrie VI, Gloria IX 

Mode VIII Kyrie I, Gloria III, Sanctus IV 

185. It has already been said (par. 4) that any modern key may 
be chosen for a Gregorian melody provided the Gregorian intervals 
are respected. The melody will always be the same if the intervals 
are retained: the change of key changes only the pitch, not the 
melody. The Kyrie of the Requiem Mass, for example, may be sung 
in any key; it is reproduced here in four different keys. 


1 J J i ' n 




Ky - ri - e e - 


le - i - son. 

Fig. 112 



186. To find a convenient key for any Gregorian melody the 
following procedure may be used : 

(1) Find the highest sol-fa note of the melody; (2) Make that 
note the highest piano note which the choir can sing with ease;. 
(3) From that note determine what piano note is do of the sol-fa 
notes. That note will be the key for the Gregorian melody. As a 
specific example, this procedure can be applied to the Requiem 

(1) The highest note is fa; 

(2) The highest note the choir can sing comfortably is E, 
let us assume; accordingly E is fa; 

(3) If fa is E, then sol is F#, la is G#, ti is A# and do is B. 
The key to be used, therefore, is the key of B: 


fa sol la ti do re mi fa 

Fig. 113 

If the key of B is found to be too difficult to play, the entire melody 
may be lowered a half-tone to B? or raised a half-tone to C: 







fa sol la ti do re mi fa 

fa sol la ti do re mi fa 

Fig. 114 

Another practical scheme is based on the dominant. Since the domi- 
nant of the mode is practically the middle of the mode and since the 
notes A or B? are practically the middle notes of the voice, either of 
these two notes may be chosen for the dominant of the mode to be 



11 187 

187. The following figure gives a convenient key for each mode. 

Mode I (re - re) 

Mode II {la -la) 

IS jjjJJ^rrnjj ^fc 

234567 (X) 12 67123456 

Mode III (mi - mi) 

Mode IV (ti - ti) 

3 4 5 6 7 1 2 3 

7 12 3 4 5 6 7 

Mode V (fa - fa) 



Mode VI (do - do) 


4 5 6 7 (X) 1 2 3 4 


12 3 4 5 6 7 1 

Mode VII (sol - sol) 

Mode VIII (re - re) 

p jjjjJJJriiljjJJJJrr 

5 6 7 12 3 

2 3 4 5 6 7 12 

Fig. 115 

NOTE: It is superfluous and impractical to include a series of exercises 
here. Any division taken directly from the chant to be studied can be used as a 
vocal exercise. It seems more serviceable for the choirmaster to lead his singers in 
reviewing and repeating more difficult passages practiced as vocalises. In this 
way the voice is exercised and at the same time the melody is definitely impressed 
upon the memory of his singers. Thus, for instance, if the melody on the words 
et semini ejus of the Requiem Offertory presents a difficulty (as it usually does), 
the passage should be repeated again and again on the sol-fa notes. After the 
melody is memorized in this way, the passage can be sung to the various vowel 
sounds. When a rhythmic difficulty arises, the notes can be sung in groups until 
the rhythm is felt. Finally, the notes should be sung to their proper text and 
proper rhythm. This method appears to be more practical and time-saving than 
the use of standardized exercises. Of course, at the beginning simpler passages 
are recommended. No melody more simple than the Requiem Kyrie can be found. 


If 188-189 


188. The psalms comprise the major portion of the Vesper 
service. This beautiful and rich liturgical devotion of the Church 
has unfortunately been permitted to fall into disuse. The reason 
may be the same as dealt the death blow to the chant at large: the 
introduction of melodies too difficult and too florid to be sung by 
the laity. Parishes were forced to resort to the use of professional 
singers and consequently forced to bear a heavy financial burden. 
Since most of the parishes were unable to support such a choir, the 
Vespers were discontinued, becoming a memory of the past. As an 
aid and encouragement to the restoration of this service, therefore, 
this chapter on psalmody is included in this book. Beginners should 
not be discouraged by the intricacies of psalmody: it is less difficult 
to sing a service than to explain it. Furthermore, the details ex- 
plained here should not be committed to memory since the Liber 
clearly prints the notation, cadences and rhythmic nuances. After 
a little practice this service can readily be mastered. 

189. Vespers are sung in the manner given below. Off. refers to 
the Officiant; C. to the Choir or Congregation; Ch. to the Chanter. 1 

I The Pater and Ave are said silently. 
II Off. Deus in adjutorium. C. Domine, ad adjuvandum. 
Gloria Patri, Sicut erat and Alleluia or Laus tibi, Domine. 

III The five psalms with their antiphons. 

IV Off. Chapter (Capitulum). C. Deo gratias. 

V Hymn with proper versicle and response (designated by 

VI Antiphon to Magnificat, Magnificat and repetition of 
VII Off. Dominus vobiscum. C. Et cum spiritu tuo. 
VIII Off. Oration of the day. C. Amen. 
IX Commemorations, if any. 
X Off. Dominus vobiscum. C. Et cum spiritu tuo. 
XI Ch. Benedicamus Domino. C. Deo gratias. 
XII Off. Fidelium animae. C. Amen. 

XIII Pater noster silently. 

XIV Off. Dominus det nobis suam pacem. C. Et vitam aeternam. 

XV C. The proper anthem of the Blessed Mother with versicle 
and response. Off. Oration. C. Amen. 
XVI Off. Divinum duxilium. C. Amen. 

! A11 references given apply to the Liber Usualis, edition No. 801, published 
by Desclee, Tournai, Belgium. 


PSALMODY if 190-194 


190. The congregation recites the Pater noster (Our Father) 
and Ave (Hail Mary) silently with the officiant, after which he 
intones the Deus in adjutorium in either the festal or the solemn 
tone as given on page 250 of the Liber. The choir replies with the 
Domine, ad adjuvandum, the Gloria P atria, and the Sicut erat. The 
Alleluia is added on all Sundays during the year except from 
Septuagesima Sunday to Palm Sunday inclusive. On these days the 
Laus tibi, Domine is substituted. 


191. The officiant then intones the first antiphon to the asterisk. 
On all double feasts the antiphons are sung in full before and after 
each psalm. On semi-double and lesser feasts, however, the antiphon 
preceding the psalm is sung to the asterisk; after the psalm it is 
sung in full. On double feasts, therefore, the officiant intones the 
first antiphon to the asterisk and the choir continues it. On semi- 
double and lesser feasts, after the officiant has intoned the antiphon 
to the asterisk, the chanter intones the first verse of the psalm. The 
first section of the choir joins at the second half and completes the 
verse; the other half of the choir sings the second verse. The two 
sections alternate thus up to and including the Sicut erat, after 
which the chanter and both choirs sing the antiphon in full. The 
ordinary Sundays of the year are semi-double. 

192. When the words of the antiphon are identical with the first 
words of the psalm, they are not repeated but are fused into the 
psalm. So, for instance, the first words of the first antiphon for or- 
dinary Sundays are Dixit Dominus and the first psalm begins Dixit 
Dominus Domino meo. When, therefore, the officiant has intoned 
the words Dixit Dominus of the antiphon, the chanter continues 
Domino meo and the first half of the choir joins him at Sede a 
dextris ; the second half then sings the second verse, the first half the 
third verse, etc. 

193. The chanter intones all the other antiphons and the first 
half of the first verse. The two sections of the choir (or congre- 
gation) then alternate throughout the psalm including the Gloria 
Patri and Sicut erat. All repeat the antiphon in full after each psalm. 
The Vespers for Eastertide have but one antiphon (see par. 215). 

194. The antiphon to the psalm may be written in any one of 
the eight modes. The psalm tone, however, always follows the mode 
of the antiphon. If the psalm tone has more than one final cadence 
(par. 198), that cadence is chosen which will best join the end of the 
psalm to the beginning of the antiphon after the last verse. It can 
readily be concluded from this that the liturgy intends antiphon 
and psalm to be considered and rendered as a complete entity. 


c 195-198 



195. The first verse of each psalm is composed of: [a) an in- 
tonation; (b) a reciting tone; c a mediant cadence; d a reciting 
tone for the second half : e a final cadence- Longer verses are broken 
by (f) a flex. All other verses of the psalm follow the same pattern 
except that they do not have an intonation (see pages 112 to 117 
of the Liber). 

196. The intonation. The intonation forms the connecting link 
between the last note of the antiphon and the reciting note of the 
psalm. It is used for the first verse only; all succeeding verses begin 
with the reciting tone. The intonation occurs on the first syllables 
of the words regardless of their verbal character. The rhythm of the 
intonation will follow the nature of the syllable. Example: 


Int. Reciting tone 

Ma - gna o - pe - ra Do-mi-ni. Con - fi - te -bor ti-bi 
Antiphon Int. Rec. tone 

t ■ « I ' 


• 1 ■ 

Al - le - lu - ia. Di - xit Do-mi-nus 

Fig. 116 

The intonation consists of one note and a neume in Modes 1 I,, III. 
IV and VI; of three single notes in Modes II. Y and YIII; of two 
neumes in Mode VII. 

197. The reciting tone. The reciting tone is the dominant of 
the mode in which the psalm is written. In Modes III and VIII 
it is do, the modern dominant; in Mode IV it is la. It is the same in 
both parts of the verse except in the Tonus Peregrinus which will 
be considered separately par. 213). All the verses of the psalms. 
except the first, begin on the reciting tone. 

198. The mediant and final cadences. A cadence is a melodic 
formula that leads the various parts of a versicle to res:. Each 
psalm verse is divided into two parts, the first ending in the mediant 
and the second in the final cadence. All the modes except VI have 
but one mediant cadence. But all except Modes II. Vand VI have 
more than one final cadence. When a psalm tone is given, its final 

; The Liber refers to the Modes as Tones. In this chapter the terms 'tones' 
and 'modes' are to be considered synonymous. 


PSALMODY 1f 199-202 

cadence is designated by either a small or a capital letter of the 
alphabet. The use of the small letter indicates that the note repre- 
sented is not the final of the mode (par. 174). A capital letter sig- 
nifies that the note on which the cadence ends is also the final of 
the mode in which the psalm tone is written. The letters used re- 
present the sol-fa notes in this order: 




















In the last chapter (par. 174) it was stated that there are only four 
finals. These may be represented by the letters D, E, F and G (in 
sol-fa notes: re, mi, fa and sol). But since another scale position is 
at times used for Mode IV, the letter A is included as the final of 
that mode. 

199. If a psalm tone has more than one cadence on a certain 
note, the tone to be chosen is specified by the addition of a small 
number beside the letter, as: D 2 , a 3 . 

200. An asterisk added to a letter (as A* or G*) indicates that 
the final cadence is extended to the note above it. These cadences 
are quite rare for they are only used in two modes: the A* in Mode 
IV and the G* in Mode VIII. The first (A*) indicates that the 
cadence ends on la (6) and is extended to ti (7); the second G* 
indicates that the cadence ends on sol (5) and is extended to la (6). 

201. In designating the psalms to be used for a proper Vespers, 
the Liber gives: (1) the titles of the psalm (its first words); (2) the 
tone to be used by a number representing one of the eight modes; 

(3) the cadence (final) by a small or capital letter of the alphabet; 

(4) the page on which that psalm is to be found. For example, we 
may find these designations before the antiphons of proper Vespers: 

Psalms: 1. Dixit Dominus. l.g.pAIS. — 2. Confitebor. 
3a 2 . p. 136. — 3. Beatus vir. 4. A*. £.144. — 4. Laudate pueri. 
l.a 3 . £.148. —5. In exitu. 8.G. £.159. 

This means: The first psalm is the Dixit Dominus in Mode (or 
Tone) I with the cadence on sol (which is not the final) as found on 
page 128. The second psalm is the Confitebor in Tone III with the 
second of the la endings as found on page 136. The third psalm is 
the Beatus vir in Tone IV with the ending on la which is extended 
to ti. The fourth psalm is the Laudate pueri in Tone I with the 
cadence on the third of the la endings as found on page 148. The 
fifth psalm is the /;/ exitu in Tone VIII with the cadence on sol 
which is the final of that Mode as found on page 159. 

202. The cadences always center around the last or the last two 
accented syllables of the verse. Hence there are two kinds of 
cadences: a cadence of one and a cadence of two accents. A cadence 
of one accent may be preceded by one. two or three preparatory 


H 203-204 


203. The accents affected by the cadence are printed (in the 
Liber) in heavy type: the preparatory syllables are printed in 
italics. It may happen on infrequent occasions that because of a 
peculiar disposition of the words the cadence will occur on a sylla- 
ble that is not accented. In a cadence of one accent, for example, 
provision is made for the extra syllable between the accent and the 
last syllable in dactylic words (par. 97). 

b , 




■ • 


se - 
no - 


e - 

mi - 



Fig. 117 

This extra note is printed as an open note in the chant books. In a 
cadence of two accents, two open notes are supplied for the extra 
syllables in dactyls: 





■ • 

• ■ 





i - 

ni - 














se - 












(a) 2 spondees 

(b) 2 dactyls 

(c) 1 spondee and 1 dactyl 

(d) 1 dactyl and 1 spondee 

Fig. 118 

204. Monosyllabic words that occur within the cadence convert 
spondees into dactylic cadences (Fig. 119 [a]) and move the place 
of the cadence in dactyls (Fig. 119 [b]) : 





k a 









Fi - 
Si - 



(a) 1 spondee and 1 monosyllable 

(b) 1 dactyl and 1 monosyllable 

Fig. 119 


H 205-208 

It should be noted that although the last syllable of the word 
Dominus in (b) is affected by the cadence, the syllable itself is 
not stressed. The melody but not the nature of the last syllable is 
changed. As a last syllable it must be rendered softly. 

205. Provisions are made for two syllables between two accents. 
If, however, more than two syllables occur between them, the ca- 
dence is shifted to another syllable: 



c ■ 




more syllables 6r- 




cu - 



cu - 

se - 


Fig. 120 

206. Each half of the verse ends in a cadence; the first half 
closes in the so-called mediant cadence, the second half in the final 
cadence. Each half possesses its own climax to which the rhythm 
rises in dynamics and agogics and from which it falls to rest. Since 
the mediant cadence is the end of only half the verse, it is less re- 
tarded than the final cadence. But since the first half is a complete 
phrase, there must be a definite pause of at least two pulsations 
before the second half is begun. When sections of choirs alternate, 
there seems to be no need for a pause between the verses. In order 
to avoid giving an impression of haste in singing, especially since 
each beginning is active, it is well for the singers to exaggerate the 
pause after the mediant cadence. When Vespers are sung in churches 
with prolonged echoes, the pause between the two halves of the 
verses might even be extended to three or four pulsations. The pause 
suggested is important; the lack of all pause will convey the idea 
of a hurried and rushed service. Activity at the beginning of all 
verses avoids dragging; a restful pause at the halves prevents rush- 
ing. An asterisk appears at the mediant cadence in the text of the 

207. A cadence of one accent is found in the mediant of Tones 
II, V and VIII. The accent of a spondee will always occur on a 
rhythmic rise note and the accent of a dactyl will always occur on 
a rhythmic fall note. In this way the spondaic accent is always light 
and brief; the dactylic is always stressed (pars. 106, 107). This 
rhythmic arrangement is assured by the addition of the open note 
in dactylic cadences, (see Fig. 117.) 

208. A cadence of two accents occurs in the mediant of Tones 
I, III and VII, and in the final of Tones V and VII. Since the first 
of the accents effected by the cadence is always higher in melody 
than the second, the first should be considered the climax of the 
verse. The spondees and dactyls are treated as in the former ca- 
dence. Figure 119 presents an example of the arrangement of syl- 


1f 209-212 


209. A cadence of one accent with one preparatory syllable 

occurs in the final of Tones II and III (first form) and in the mediant 
of Tone VI. The syllable before the last accent is affected regardless 
of its verbal character. 

rec. tone 

1 pr. 



if* m 

% * 




















Fig. 121 

210. A cadence of one accent and two preparatory syllables 

occurs in the mediant of Tone IV and in the final of Tones I, III 
(second form), VI and VIII. The two syllables before the accent 
are affected by this cadence. The preparatory syllables of tone IV 
(mediant) and Tones I and VIII (final) consist of two single notes; 
of Tone III of two neumes, of Tone VI of one note and one neume. 
The notes or neumes are accorded a rhythm in keeping with the 
nature of the syllable they adorn. 

1 ' — ■ 

rec. tone 

1 pr. 

2 pr. 



5 . 

■ , , 












mi - 
















Fig. 122 

211. A cadence of one accent and three preparatory syllables 

occurs in the final of Tone IV. This is perhaps the most intricate of 
all the cadences not only because of the three preparatory syllables 
but also because of the special form of the dactylic cadence. 

212. The special form of the dactylic cadence is described in 
the Liber as 'an extra note added in anticipation of the accent in 
dactylic cadences'. In the accent-cadences as described above, the 
accents of spondees are always light and those of dactyls always 
stressed (par. 207). In this special form of cadence, however, the 
rhythm of the accent is reversed : the accents of spondees are always 
stressed (because they occur on the neume), and the accents of dactyls 
are always light (because they occur on the isolated note before the 




neume). This effect is produced by the addition of an open note which 
is used only on the accents of dactyls. This type of cadence was anti- 
cipated in the discussion of the effect of the accent on a rhythmic 
rise and fall note (par. 105). To master this cadence the singer need 
only remember that the accent of every spondee is sung to the 
neume (and is therefore stressed), that the accent of dactyls is sung 
to the open note (and is therefore light and brief), and that the 
syllable between the accent and the last in a dactyl is sung to the 
neume. The Liber places a bracket over the open note and the 
neume; it prints both the accent and the syllable between it and 
the last in heavy type. This special form of cadence occurs in the 
mediant of Tone III, in the. final of Tone I (D 2 ending) and IV 
(E ending). 

Tone I (D 2 ) 

1 accent and 2 

preparatory syllables 

/ ac. 


/ pr. 

2 pr. 

G a 





















* — 




Tone III 
2 accents 

/ ac. 

2 ac. 




\r - 























Fig. 123 



Tone IV (E ending) 
1 accent and 3 preparatory 

(one of which is a neume) 

/ ac. 


f pr. 


3pr. t a 

5 . 







■ • 































Fig. 123 (cont.) 

213. The Tonus Peregrinus is a special psalm tone that is used 
for the psalm In exitu which, however, uses other psalm tones too. 
The psalm In exitu is in fact a combination of two psalms, both of 
which commemorate the return of the children of Israel from cap- 
tivity. Hence the name Tonus Peregrinus or 'Pilgrim Tone'. The 
intonation consists of a neume {la - te) to be used on the first syllable 
of the first verse only; the reciting tone of the first half is /a, of the 
second half, sol. Both cadences supply an extra (open) note for the 
syllable between the accent and the last of the dactyls. 

214. Longer verses are broken by a flex (Latin for 'bend'). 
This breaking always occurs within the first half of the verse and is 
designated by a dagger (f) in the Liber. The flex may descend either 
a whole tone or a minor third, but never a half tone. If the re- 
citing tone is do or fa, the flex does not descend a half tone to ti or 
mi respectively, but rather a minor third to la or re. But when the 
reciting tone is any other note, the flex descends a whole tone. Fur- 
thermore, the flex descends on the syllable after the accent: on the 
last syllable of a spondee and on the last two of a dactyl. For this 
reason an open note is added for the extra syllable after the accent 
in dactyls. 


whole tone 

minor third 



« • 

A a 

" " Q ■• 

T* " « 

.su - 6 - rum f 
com-mo-dat f 

Do - mi - no f pau - pe - ri - bus t 

De - o t 

Fig. 124 

PSALMODY 1f 215-217 


-■ — m- 

i - ra - see - tur t 
vi - vi - fi - ca - bis me t 

Fig. 124 (cont.) 

Since the flex is only a breaking or 'bending' of a longer verse, there 
should be no pause between it and the remaining half of the verse. 

215. Vespers for Eastertide differ from the ordinary Sunday 
Vespers in that the five psalms are sung to one antiphon. The anti- 
phon consists of a three-fold repetition of Alleluia. But since on 
semi-double feasts the antiphon preceding the psalms is sung to the 
asterisk, one Alleluia is sung before the five psalms and the whole 
antiphon (of three Alleluias) is sung after the psalms. After the 
officiant intones the antiphon, the chanter intones the first psalm 
{Dixit Dominus) in Tone VII to correspond to the mode of the 
antiphon. This tone consists of an intonation of two neumes to be 
sung on the first two syllables of the first verse only. All the other 
verses of all the psalms begin on the reciting tone. There should 
be no interruption between the Sicut erat of one psalm and the be- 
ginning of the next. Each section of the choir sings alternate verses 
throughout the five psalms. After the last Sicut erat the entire choir 
sings the antiphon in full. 

216. Vespers for the four Sundays of Advent have proper anti- 
phons and proper psalm tones to fit them. They are clearly indicated 
in their proper places after the Mass for these Sundays (see page 
323 of the Liber for Vespers for the First Sunday of Advent). After 
the proper antiphons, the Liber gives the final cadence under the 
letters Euouae. These letters represent the vowel sounds of the 
words saeculorum. Amen with which all psalms end. Thrs feature 
has two practical values: (1) When the psalms are recited from mem- 
ory (as they generally are in monasteries), the final cadence is given 
to save the singers the necessity of looking for the cadence under 
the psalm tone. (2) Since the cadence is printed after the antiphon, 
the relation of the last notes of the psalm to the first of the antiphon 
is more readily seen. 


217. After the repetition of the last antiphon the officiant sings 
the Chapter (or Capitulum) according to the tone given on page 123 
of the Liber. The reciting tone is do, the flex to /a, a mediant cadence 
of one accent and two preparatory syllables, a final cadence of one 
accent. The last syllable is sung to the dotted notes sol - la. The 
choir replies as follows: 


If 218-223 GREGORIAN CHA7v[T 

6 5 5. 6. 
gra-ti - as 

Fig. 125 


218. The officiant intones the first line of the proper hymn and 
the choir continues it. For ordinary Sundays of the year the Lucis 
Creator optime is sung in any of the three tones, as given on pages 
256 - 259 of the Liber. On other Sundays, the particular hymn is 
clearly indicated. After the hymn, the chanter sings the proper ver- 
sicle and the choir answers with the proper response. The versicle 
and response may be sung in either of two tones as given on page 
118. On the greater feasts of the year they have a more ornate tone 
as given in their proper places; for example, the Vespers for Christ- 
mas, Versicle and Response on page 413. 


219. After the versicle and response the officiant intones to the 
asterisk the antiphon to the Magnificat, which is always proper. 
On semi-double feasts the chanter begins the Magnificat after the 
intonation of the antiphon; on double feasts, however, the choir 
continues and sings the antiphon in full before the Magnificat. 

220. The Magnificat is like a psalm tone. But there are differ- 
ences. Each tone has two forms, a simple and a solemn, as indicated 
in the Liber after the antiphon to the Magnificat. The former is 
used on ordinary Sundays; the latter may be used on principal 
feasts, that is, on doubles of the first and second classes. Like the 
psalms, the Magnificat follows the mode of the antiphon. The in- 
tonation is repeated at the beginning of each verse of the Magnificat. 


221. After the antiphon to the Magnificat is repeated, the offi- 
ciant sings Dominus vobiscum, to which the choir responds with the 
usual Et cum spiritu tuo. 


222. The officiant then sings the oration of the day, which is 
always proper. The choir replies Amen. 


223. The necessary commemorations to be made may be foiind 
in the Ordo. They are made as follows. The antiphon to the Magni- 
ficat of the feast is sung in full by the choir after being intoned by 


PSALMODY 1224-229 

the chanter; the chanter sings the versicle and the choir replies with 
the response. Both the versicle and the response to the commemora- 
tions are sung in the simple form, that is, a descent of a minor third 
after the last accented syllable, as shown on page 118 of the Liber. 
The officiant adds the usual conclusion only to the last oration of 
the commemorations. Other commemorations are made in like 
manner immediately after the oration, but without the usual con- 
clusion. To the last oration the officiant affixes the conclusion Per 
Dominum nostrum or Qui vivis et regnas to which the choir replies 
Amen. If there is no commemoration of a double feast or of an 
octave, the Suffrage of the Saints is sung on all Sundays of the year 
except during Advent and Passiontide. During Eastertide, however, 
the 'Commemoration of the Cross' is used instead of the Suffrage. 
These are given on pages 260 and 261 of the Liber. 


224. Then follows the Dominus vobiscum, sung by the officiant, 
to which the choir replies Et cum spiritu tuo. 


225. The proper Benedicamus Domino is sung by the chanter, 
and the choir replies with the corresponding Deo gratias as indicated 
on pages 124 - 127 of the Liber. 


226. The officiant then says, recto tono, in a low voice, Fidelium 
animae, etc., as on page 261 of the Liber. The choir replies at the 
same pitch; Amen. 


227. If Compline does not follow, the Pater is said silently. 


228. After the Pater the officiant adds Dominus det nobis suam 
pacem to which the choir replies Et vitam aeternam. Amen, in the 
same tone as the Fidelium, as on page 261 of the Liber. 


229. The officiant then intones the anthem to the Blessed Virgin 
according to the season. These anthems are in two tones: the simple 
and the solemn. Either may be used at the discretion of the director, 
according to the ability of the choir or at the request of the pastor. 
From the first Sunday of Advent to the Second Vespers of the Feast 
of the Purification (which may occur on a Sunday), the Alma Re- 
demptoris Mater is sung (solemn tone, page 273; simple, page 277). 
During Advent the versicle and response are Angelus Domini and 


If 230-231 


Et concepit, as on page 274. From Christmas to Second Vespers of 
the Purification the versicle and response are Post partum and Dei 
Genitrix, as on page 274. From the Feast of the Purification to 
Palm Sunday the anthem Ave Regina coelorum (solemn tone, page 
274; simple, page 278) is sung. The versicle and response are always 
the same as on page 275. From Easter Sunday to Pentecost the 
Regina coeli (solemn tone, page 275; simple, page 278) is sung. 
From the feast of the Most Holy Trinity to the -first Sunday of 
Advent the Salve Regina (solemn tone, page 276; simple, page 279) 
is sung. The versicle and responses to all the anthems are sung in 
the simple form (see page 118 of the Liber). 


230. After the officiant has chanted the proper prayer to the 
anthem' and the choir has answered Amen, the officiant says recto 
tono and in a low voice Divinum auxilium, etc., to which the choir 
replies Amen, as on page 261 of the Liber. Thus Vespers are con- 
cluded. Whatever else is added, such as Benediction, is optional. It 
may be well occasionally not to add anything to the Vespers, first, 
because additions are not prescribed in the liturgy; secondly, be- 
cause nothing else need be added. 

231. Not all the psalm tones can be sung to the same modern 
key. If, for example, the key of C were chosen, the low la of Mode 
II would be the A below middle C, and the high sol of Mode VII 
would be the G above the modern staff. It is suggested therefore 
that a definite note, either A or Bb be chosen as the reciting note of 
all the psalms. Fig. 126 gives the transcription of the first half of 
the first verse of each psalm for Sunday Vespers, with the note A 
as the reciting tone for each. 

Tone 7 

s ■ . = — d— m — = — a — ■— 

■ ■ 

Dixit Dominus Domino me- o: * 

i'j,Mhj W.MJ 

Fig. 126 


m HHM — ■ — D — ■—£>- 

Tone 3. 

Confitebor . . toto cor- de me • o : * 

b b h b h 

* 9 w fi tat 


-■ — ■ — » 

-q — m- 

Tone 4, 

Be - a - tus vir qui timet D6-mi-num : * 


d) * J* J J J ; J J *4#) * 


* ■ ■ • ° ■ 

Ton^ 7. / Lau - da - te pii - e - ri D6-mi-num : 

\ -tea 

H a ■ ■■ ■ ■ , bi , a : 


In exitu Isra £/ de Ae - gy - pto 


Fig. 126 (cont.) 

If 232-238 


232. The Latin language, like any other language, contains two 
sounds: vowel and consonant. A vowel sound is produced in such a 
manner that the air is allowed to pass in a continuous stream over 
the vocal chords and through the mouth without any obstruction 
which might cause audible friction. A consonant sound is one that is 
produced by either a partial or total obstruction of the air passage. 

233. There are five vowel sounds in Latin, represented by the 
letters u, o, a, e and i. This number should be compared with the 
eighteen vowel and diphthong sounds in English. The letters which 
represent the vowel sounds in Latin differ from the English in that 
each letter designates a single, pure and invariable sound. In Eng- 
lish a vowel may have various sounds: for example, the sound of a 
is different in the words father, all, at, rare. In Latin, on the other 
hand, a is always a in father. 

234. In the formation of the Latin vowel sounds two organs of 
the mouth function : the lips and the tongue. The tongue may assume 
a position low or high, in front or in back. The lips may be fully 
open, partly open, or closed, either in a rounded or oblong position. 

235. The most open of all sounds is the vowel sound a. The Latin 
a is pronounced like the a in father, calm, far, garden. The tongue 
lies low in the mouth and the lips are" fully open. Examples: al-ma 
(ahl-mah), san-cta (sahn-ktah), ad (ahd), pax (pahks). 

236. If the tongue is raised slightly at the back of the mouth 
and if the lips are closed slightly in a rounded or oval position, the 
sound of the Latin k o can be produced. This letter represents the 
vowel sound in the words saw or law. There is no u sound in this 
vowel as there is in the English no (no-oo). Examples: san-cto (sahn- 
ktaw), a-do-ro (ah-daw-raw) , bo-no (baw-naw), tan-to (tahn-taw), 
ho-san-na (aw-sahn-nah). 

237. If the tongue is again raised at the back and the lips more 
definitely rounded, the sound of the Latin u is produced as the 
vowel sound in food but never as in foot. Examples: tu-am (too-ahm), 
cu-jus (koo-yoos), mun-dus (moon-doos), so-las (saw-loos), sunt 

238. If we begin from the vowel sound of a, raise the tongue 
slightly to the frorit, and close the lips slightly in an oblong position, 
we can utter the sound of the Latin e. This letter represents the 



sound of the vowels in the English words red, said or bed. It should 
be remembered that this is a pure sound containing no 'glide-off' 
to i as in innumerable English words: make, ale, name, lace, etc. 
(meh-eek, eh-eel, neh-eem, leh-ees). Examples: et (eht), De-us 
(deh-oos, not deh-ee-oos), se-des (seh-dehs), ple-na (pleh-nah), 
Jesus (yeh-soos), and also Ky-ri-e (kee-ree-eh, not kee-ree-eh-ee). 

239. If we raise the tongue farther forward and close the lips 
in a more oblong position, we can produce the Latin vowel sound of 
i as in the English words see, be, feed, meet. Chris-te (kree-steh), 
i-ni-mi-ci (ee-nee-mee-chee), ho-mi-ni-bus (aw-mee-nee-boos), no- 
bis (naw-bees), pec-ca-tis (pehk-kah-tees), ji-li-i (fee-lee-ee), in 

240. Two vowel sounds combined form diphthongs (from the 
Greek meaning 'double-sound'). Every vowel of a Latin word forms 
a separate syllable except when the vowels are joined in a diphthong. 
So, for instance, the eu of Deus are two separate vowels, each form- 
ing a distinct syllable: De-us; but the eu in euge is a double vowel 
forming only one syllable: eu-ge. These two vowels are pronounced 
in the time value of one syllable in speech but in singing, one vowel 
must be held and the other inserted quickly. The principal vowel 
is held and the second is either a 'glide-on' or a 'glide-off' vowel. An 
example from our own language will elucidate. When the word old 
(aw-oold) is sung, the principal vowel aw would be sustained for 
the length of the note and the secondary oo inserted before passing 
to the next vowel. The oo in this case is a glide-off vowel. In the 
word new, however, there are also two vowel sounds, the first being 
a glide-on and the other the sustained : neeoo. Hence before the vowel 
oo is sung, the ee is quickly inserted before it. These double vowels 
occur less frequently in Latin than in English. Very often a liques- 
cent (par. 43) is used to facilitate the singing of the extra vowel. 

241. There are five diphthongs: ae, oe, au, eu, ei. Under certain 
circumstances the consona'nts j, y and g representing an i sound are 
added to this number. 

242. The diphthongs ae and oe are pronounced like the simple e. 
Examples: cae-lo (cheh-law), poe-na (peh-nah), sae-cu-li (seh-koo- 
lee). In the diphthong aw the first is the sustained or principal vowel, 
the second the glide-off. If, therefore, a melody of five notes is to be 
sung to this syllable, the five notes are sung to the a and the u is 
inserted before the next syllable. The principal vowel is bold face. 
Examples: au-di-vit (ahoo-dee-veet) , Mau-rus (mahoo-roos) . 

243. There is a dispute in regard to the correct pronunciation 
of the Greek diphthong eu. Some allow the first, others the second 
vowel to be the sustained portion of this diphthong: the u is the 
glide-off in the former opinion and the e the glide-on in the latter 
opinion. Still others give it the sound of oi as in toil. It is recommend- 
ed, however, that the u be accepted as the sustained sound and the 
e as the glide-on sound. Examples: eu-ge (ehoo-jeh), seu (sehoo), 
eu-cha-ri-sti-a (ehoo-kah-ree-stee-ah) . 



244. The sound ei is a diphthong only in the interjection hei 
which is pronounced ehee; otherwise the combination forms sepa- 
rate vowels and syllables. Examples: re-i (reh-ee), De-i-tas (deh- 
ee-tahs). These are not diphthongs. 

245. The j, given the sound of i or y, forms a diphthong with the 
vowel which follows it. Examples: eius or ejus (eh-eeoos, pronounced 
eh-yoos), al-le-lu-ia or al-le-lu-ja (ahl-leh-loo-yah), ma-je-sta-tis 
(mah-yehs-tah-tees), cu-jus (koo-yoos). 

246. The y following a vowel forms a diphthong with that vowel. 
The vowel is always the principal sound of the syllable, the y form- 
ing a glide-off from it. Examples: Ray-mun-dus (rahee-moon-doos). 
When it appears between two consonants, however, it forms a 
separate syllable and is pronounced like i. Examples: Ky-ri-e 
(kee-ree-eh), mar-ty-ri-um (mahr-tee-ree-oom) , cym-ba-lum (cheem- 
bah-loom) . The Hebrew name of Moses is an exception : the y forms 
a separate syllable although it follows a vowel: Mo-y-sis (maw- 

247. The g when followed by n is pronounced like y, thus form- 
ing a diphthong with the following vowel. Examples: a-gnus (ahn- 
yoos); si-gni-fer (see-nyee-fer) , ma-gni-fi-cat (mah-nyee-fee-kaht). 
The g when followed by u and another vowel is also converted into a 
diphthong sound. Examples: lan-gue-o (lahn-gooeh-aw), san-guis 

248. The following principles and suggestions for the correct 
pronunciation of the vowel sounds should be observed : 

1. Every Latin vowel is pure; that is, it has but one sound. 
Care must be taken that English diphthongs do not influ- 
ence the pronunciation of Latin vowels (pars. 238, 240). In 
English we customarily close the o sound with an u — (no as 
no-oo) and an e sound with an i {made as meh-eed). The last 
o in Domino is exactly like the first (daw-mee-naw). The 
difference between Dei and de should be clear (deh-ee and 
deh). The Latin word meo has but two vowel sounds: eh 
and aw. We English-speaking people would like to insert 
two alien sounds: oo and ee, thus pronouncing the word 
meh-ee-aw-oo instead of meh-aw. Between the two words 
Kyrie and eleison there must be no ee sound: kee-ree-eh 
eh-leh-ee-sawn not kee-ree-eh-ee eh-leh-ee-sawn. Overcoming 
this difficulty will not only simplify but also beautify the 
rendition of the chant. It will simplify it, for difficult foreign 
sounds can not be rendered correctly and smoothly; it will 
beautify it, for Latin when pronounced purely is very melo- 
dious in its wealth of pure and musical sounds. 

2. An i sound should never be introduced before a u unless 
definitely printed in the word as a y or a j. Sae-cu-la is pro- 
nounced seh-koo-lah, not seh-keeoo-la; do-cu-men-tum is pro- 
nounced daw-koo-mehn-toom, not dok-keeoo-men-tum. 



'3. The succeeding consonant must never interfere with the 
purity of the vowel. The purity of the i must not be disturbed 
by the 5 that follows it: no-bis is pronounced naw-bees; 
san-ctis is pronounced sahn-ktees ; Chris tus is pronounced 

249. In Latin there are eighteen consonants: b, c, d, f, g, h, k, I, 
m, n, p, q, r, s, t, v, x and z. A consonant is a sound that is produced 
by an appreciable interference of one or more of the vocal organs 
with the free flow of air. This contrasts with a vowel sound wherein 
the flow of air is unimpeded. The organs that can interrupt the flow 
of air are the tongue, the teeth, the lips or the glottis, or any com- 
bination of them. 

250. The consonants are generally grouped into two classes: 
those affected by the continuity of the flow of air; those affected 
by the use or non-use of the vocal chords producing the sound. 

251. Those consonant sounds which totally stop the flow of air 
and suddenly release it are called explosives. The explosives are 
p, t, k, b, d and g. The remaining consonant sounds:/, /, m, n, r, s 
and v are sustainable in the sense that they can be prolonged with- 
out the stopping of the flow of air. 

252. When consonants are classified as affected by the use or 
non-use of the vocal chords, they are called voiced and unvoiced. 
The voiced consonants, produced with the aid of the vocal chords, 
are: b } d, g, v, /, m and n. The unvoiced consonants, produced without 
the aid of the vocal chords, are: p, t, c,f, s, x and z. 

253. The consonants, therefore, may be classified as follows: 

explosive sustainable 

voiced: b, d, g v, I, m, n, r 

unvoiced: p, t, c f, s, x, z 

The voiced sustainable consonants are often sung to liquescent 
notes (see par. 43). 

254. The Latin consonants are generally produced like their 
English equivalents. There are several notable differences, however, 
which will be discussed below. 

255. The letter c is pronounced as k or as the c in car except when 
followed by the sounds of e and i. The sounds of e and i include not 
only these vowels but also the related semi-vowels like y (par. 246) 
and the diphthongs ae and oe (par. 242). Before the sounds of e and 
i, c is pronounced like the ch in chair or Charles. Examples: ca-ro 
(kah-raw), de-fun-cto-rum (deh-foon-ktaw-room), suc-cur-ren-te (sook 
koor-rehn-teh), cul-pa (kool-pah), hu-ic (oo-eek), lu-ce-at (loo-cheh- 
aht), de-cet (deh-cheht), ju-di-ci-um (yoo-dee-chee-oom), par-ce 
(pahr-cheh), cae-lum (cheh-loom), cae-co (cheh-kaw), Cae-ci-li-a 
(cheh-chee-lee-ah), cae-ci-tas (cheh-chee-tahs), cir-cum-cin-ge (cheer- 
koom-cheen-jeh), ci-ca-tri-ces (chee-kah-tree-chehs), cym-ba-lum 
(cheem-bah-loom), Cy-prus (chee-proos). 



256. The letter g is also pronounced in two ways: as the g in 
girl, garden or gone before all sounds except the sounds of e and i. 
Before e and i it is pronounced like the g in gender or general, rep- 
resented in the phonetic spelling by a j as in jet. Examples : glo-ri-a 
(glaw-ree-ah) , gra-tis (grah-tees), er-go (ehr-gaw), re-go (reh-gaw), 
spar-gens (spahr-jens), re-sur-get (reg-soor-jeht), re-git (reh-jeet), 
gae-sum (jeh-soom), gi-gan-tes (jee-gahn-tehs) gy-ra-tum (jee-rahr- 
toom), co-get (kaw-jeht). 

257. The letter h is not pronounced in Latin. Examples: ho- 
san-na (aw-sahn-nah), ho-sti-as (aw-stee-ahs) , hie (eek), hu-ic 
(oo-eek). The combination ph is pronounced as an/; phi-lo-so-phi-a 
(fee-law-saw-fee-ah), Phi-lip-pus (f ee-leep-poos) . The combination 
th is pronounced t: ca-tholi-cam (kah-taw-lee-kahm) , Tho-mas 
(Taw-mahs). An h following a c changes the c to a k sound even if 
the combination is followed by an e or i sound: pul-cher (pool-kehr). 

258. The Latin j is pronounced like the y in yes. Examples: 
Je-sus (yeh-soos), e-jus (eh-yoos), ju-di-co (yoo-dee-kaw) , ma-jes- 
tas (mah-yehs-tahs). For j as a vowel sound, see par. 245). 

259. The Latin q with the letter u which always fellows, is 
pronounced like kw or like the qu in quick. Examples: qui (kwee), 
quo-que (kwaw-kweh), qua-lis (kwah-lees), se-que-stra (seh-kweh- 
strah), e-quu-le-o (eh-kwoo-leh-aw). 

260. The Latin r is never silent or glided over as the r in English. 
It is always pronounced distinctly. Examples: Rex (rehks), ae-ter- 
nam (eh-tehr-nahm), cre-mer (kreh-mehr) suc-cur-ren-te (sook-koor- 
rehn-teh). Care should be taken, however, to avoid anticipating 
the r when it initiates a syllable. Example: Ky-ri-e (kee-ree-eh, not 

261. The Latin z is pronounced like ts as in cats. Examples: 
zo-na (tsaw-nah), zi-za-ni-a (tsee-tsah-nee-ah). 

262. In the combination cc before e or i, the first c is converted 
into a t sound and the second, the usual ch sound. Otherwise the 
combination is pronounced as kk. Examples: ec-ce (eht-cheh), 
ac-ci-pi-o (aht-chee-pee-aw), ac-ce-dat (aht-cheh-deht), ac-cin-ctus 
(aht-cheen-ktoos), suc-cur-ren-te (sook-koor-rehn-teh). 

263. The combination ch is always pronounced as k (see par. 
257). Examples: Mi-cha-el (mee-kah-ehl), arch-an-ge-lo (ahrk-ahn- 
jeh-law), che-ru-bim (keh-roo-beem). 

264. The combination gn is pronounced as though it were written 
ny. Examples: a-gnus (ah-nyoos), ma-gni-fi-cat (mah-nyee-fee-caht), 
i-gnis (ee-nyees), di-gnae (dee-nyeh), be-ni-gne (beh-nee-nyeh), 
co-gno-sco (kaw-nyaw-skaw) . 

265. The combination sc has the sound of sh as in she before 
e and i; otherwise as sk or the sc in scale. Examples: su-sci-pe (soo- 
shee-peh), sci-o (shee-aw), sce-na (sheh-nah), ob-scu-rum (awb-skoo- 
room), scri-ptum (skree-ptoom), in-ge-mi-sco (een-jeh-mee-skaw). 



266. The combination sch is pronounced like sk for the ch always 
retains its k sound. Examples: sche-ma (skeh-mah), schi-sma (skee- 
smah), scho-la-ris (skaw-lah-rees). 

267. The combination ti before a vowel and following any letter 
except s, t or x, has the sound of ts; otherwise it is pronounced 
naturally as ti. Examples: noc-ti-um (nawk-tsee-oom), gen-ti-um 
(jehn-tsee-oom), lae-ti-ti-a (leh-tee-tsee-ah), gra-ti-a (grah-tsee-ah), 
sen-ti-o (sehn-tsee-aw), e-ti-am (eh-tsee-ahm) ; but ho-sti-as (aw-stee- 
ahs), ex-sti-ti (ehks-stee-tee), at-ti-ne-o (aht-tee-neh-aw). 

268. The difficult word ex-cel-sis is simply ehks-shel-sees; Rex 
coe-les-tis is rehkssheh-lehs-tees. 

269. Non-Latin words are always pronounced according to the 
rule of Latin pronunciation. Examples: A-gi-os (ah-jee-aws), 
Ky-ri-e (kee-ree-eh), Sa-ul (sah-ool), a-tha-na-tos (ah-tah-nah-taws), 
i-schy-ros (ee-skee-raws) , Jo-e-lis (yaw-eh-lees), cly-pe-os (klee- 
peh-aws), Is-ra-el (ee-srah-ehl), Je-ru-sa-lem (yeh-roo-sah-lehm), 
Beth-le-hem (beht-leh-ehm), Job (yawb), a-do-na-i (ah-daw-nah-ee), 
Jor-da-nis (yawr-dah-nees), A-a-ron (ah-ah-rawn), Ae-gy-pto (eh- 
jee-ptaw), Mel-chi-se-dech (mehl-kee-seh-dehk), Br vo-kly-ni- en-sis 
(broo-klee-nee-ehn-sees) . 

270. The following practical conclusions and suggestions for 
the correct pronunciation will render the singing of the chant more 
simple, more correct and more beautiful: 

(1). Each vowel (except diphthongs) forms a separate syllable in 
the Latin word. Examples: con-su-e-tu-di-nis , de-al-ba-bor , lae-ti-ti-a, 
Mo-y-sis, A-a-ron. 

(2). Every syllable should, as much as possible, end in a vowel or 
at least a singable consonant. Examples: Ky-ri-e, al-le-lu-ia, san- 
ctus, cu-jus, do-cu-men-tum. 

(3). Consonants must never be anticipated and must never inter- 
fer with the purity of the vowel. This purity must be maintained 
and sustained for every note to be sung on that syllable. Examples: 
Ky-ri-e is sung kee-ree-eh, not keer-ree-eh; Chri-ste is sung kree- 
steh, not kris-teh; san-ctus is sung sahn-ktoos, not sank-tus; pax 
is sung pahks, not packs; om-ni-po-tens is sung awm-nee-paw-tehns, 
not awm-nip-po-tehns. 

(4). All the notes over a syllable must be sung to the pure vowel 
sound of that syllable. If there are four notes to be sung over the 
first syllabi. o f Ky-ri-e, the four notes must be sung to the pure 
vowel sound of ee. Care must be taken not to anticipate the r and 
so introduce incorrectly a sound of eh before the fourth note. Like- 
wise with the first syllable of Chri-ste. The second syllable of san- 
ctus is sung to a pure 00 sound; the second syllable of no-bis to a 
pure ee sound; the third syllable of Sa-ba-oth to a pure aw sound. 



(5V When a syllable begins with a singable consonant, the first 
note should be introduced on that sound: 




^ s^ 

■ a ■ ■ — — ps#- ■ ■ ■ - 


equi- em * lux ve-ni- et. 

Fie. 12 

,6 . When a syllable ends with a singable consonant, a fraction 
of the last note should be sung to that consonant par. 51): 



— i- ■ 

A-men. Alle-lu-ia. 

Fig:. 12 5 

Much oi the charm oi the above phrase and similar ones is lost be- 
cause of an oversight oi the nature of the groups. Group 1 is active 
ictus on accent, and beginning of word : Groups 2 and 3 are long 
dotted notes and soft .last syllable of word : Group 4 is active 
(ictus on secondary accent and beginning c: new word and should 
not be retarded by the two syllables, one on each of the group: 
Group 5 is retarded somewhat because of its nearness to the end 
while the ictus is definitely stressed melodic climax on accent of 
word • : Group 6 is definitely retarded all ends are long . This phrase 
has been inserted here as an example because its true beauty is often 
lost through a misunderstanding of the pronunciation of the words. 
Every letter of the two words, both consonants and vowels, is 

.7 . When a syllable begins with a singable consonant, that con- 
sonant must be pronounced before the vowel sound of that syllable. 


* * * * 

% ■ ■■ 

et lux perpe-tu- a 

Fig. 125 


11 270 

(8). When a phrase ends in a singable consonant, the last note is 
softly closed on that consonant. 

C > fl P ... 



glo- ri- am tu- am. 

e- le- i-son. 

Fig. 130 

In the above example, as the ah of the last syllables of gloriam and 
tuam are diminished, the m is inserted like a very soft hum. But 
when the phrase ends in an unsingable consonant, that consonant 
is pronounced so softly that it is not heard. The reason for the silence 
lies in the fact that the vowel is closed so softly that the consonant 
that follows it, even if pronounced, is not heard. If it were heard, 
the consonant would be louder than the end of the vowel. In other 
words, the dynamics, would open on the consonant after closing on 
the vowel. This feature has practical value, too, for if two or more 
members of the choir do not pronounce the consonant exactly to- 
gether, a double or prolonged sound will be heard. For example, if 
a final 5 is pronounced at different times by different members, the 
result is a series of the letter, approximating a hiss. 


-a 1 •- 

■ ■ 

so-lus sanctus. 

ve-ni- et. 

Fig. 131 



As sung by the Monks of Solesmes, final 5 at the ends of phrases 
is rarely heard. 

(9). The singing of the chant progresses from the vowel sound of 
one syllable to the vowel sound of the next. The consonants are 
inserted briskly but smoothly between them. 

(10). The vocal organs must never change their position during 
the singing of a vowel sound. In the Requiem Introit, for instance, 
the vowel sound of the second syllable of e-is (ee) is not to be changed 
for the eight notes sung to it. 

(11). Every syllable is given at least one basic pulsation (par. 47). 
This time value affords ample opportunity to pronounce every syl- 
lable distinctly. Care should, however, be exercised in the correct 
recitation of dactylic words. The syllable between the accented 
syllable and the last syllable in a dactylic word has been abbreviated 
in the course of the centuries in speech (par. 94) but it must not be 
shortened in the chant. It often happens that this syllable occurs 
in the cadence of divisions where length is always introduced. If it 
is shortened, the smooth flow to rest is invariably destroyed and it 
will seem that the rhythm 'tripped' before it fell to rest. The follow- 
ing illustration gives examples of dactylic words whose syllable 
between the accented and final syllable is shortened. 


■ • 

myste-n- um, 

gen-ti- um. 


■ • 

de - fectu - i. 

Fig. 132 






271. The Liturgy can be defined as the mystical Body of Christ 
living the life of Christ in a mystical way. It is the renewing and 
reviewing of Christ's life by the Church through the sacred cere- 
monies. Herein lies the chief difference between a liturgical and a 
non-liturgical devotion : in the former, the Church acts as a corporate 
person; in the latter, the individual acts in a personal capacity. A 
liturgical devotion is a public act; a non-liturgical devotion, private. 
Essentially, the first is performed by two or more persons acting 
together according to the rite prescribed by the Church; the other 
may be performed by one or more persons according to the tastes 
of the individual. 

272. Because the liturgical functions are public, corporate exer- 
cises, they must necessarily be performed by at least two persons 
of whom one places the petitions and conducts the rite and the other 
approves and consents to the acts. The sacred ceremonies are con- 
ducted in the form of a meeting of the faithful under the direction 
of their spiritual leader, the priest. In the Mass, in which we are at 
present particularly interested, the priest and the attendants offer 
the sacrifice together. Under most circumstances, the only par- 
ticipation of the laity with the priest at the altar is through the 
medium of the voice. This can be done either through the reciting 
or the singing voice. In the dialogue Mass the medium is recitation 
or speech; in the sung or High Mass the medium is song. For the 
former special permission of the Bishop is needed; the latter is 
prescribed, approved and highly encouraged by the Church. 

273. In the course of the centuries when the chant was gradually 
replaced by figured music, professional singers and trained choirs 
were introduced into the Church because the laity could no longer 
participate in such difficult and involved song. Much of this music 
is not only sanctioned but also encouraged by the Church. But this 
historic replacement has nevertheless relegated the faithful to the 
sad position of "silent spectators", as Pius XI calls them. The closest 
bond that the laity can form with the priest at the altar is its physi- 
cal presence in church. Beyond this the congregation can do nothing 
except, perhaps, read their missals silently and privately while the 
priest celebrates the Mass. For many centuries the Popes have con- 
stantly clamored with all the authority of their apostolic office for 


11274-277 GREGORIAN CHA^T 

a return of the laity to the sacred ceremonies. It was not, however, 
until the Motu Proprio of Pius X was published that a definite 
reform was launched. 

274. In order that the liturgical nature of the Mass and the 
people's place in it be understood clearly, the Mass must be con- 
sidered a public, community devotion. It may readily be compared 
to a civic meeting: the priest corresponds to the chairman; the 
people, to the attendants at the meeting. The priest is indeed the 
duly authorized chairman ordained by the Church to conduct the 
sacred meeting in Her name and according to Her regulations. The 
faithful as free citizens of the Church of God, duly incorporated into 
that Body through the waters of Baptism, have the right to a vote 
and a voice in the transactions. The House of God, the parish church, 
is the appointed meeting-hall. 

275. The Mass may be divided into five parts: (1) the school; 
(2) the offertory; (3) the thanksgiving; (4) the banquet; and (5) the 
dismissal. Each part (except one) begins with the priest's greeting 
to his people and the people's reply, and ends with the people's 
assent and approval to what he has done in his and their name. 

276. The priest greets his people with the familiar Dominus 
vobiscum to which the people reply, Et cum spiritu tuo. 1 The Do- 
minus vobiscum, meaning "May the Lord be with you", is similar to 
our "Good-bye", a contraction of the words "May God be with ye." 
The people return the greeting in the words Et cum spiritu tuo, 
meaning: "And with thy spirit." 

277. The people voice their approval at the end of each part of 
the Mass with the word Amen 2 , meaning "May it be so." This 
corresponds to the modern "aye" used to give consent to the chair- 
man's motion: "All in favor, signify by saying 'aye' ". The Amen 
is, therefore, the liturgical "aye". An ancient writer has said, "All 

^he reply, Et cum spiritu tuo, should always be made in a steady, joyful 
voice expressive of the holy joy and lively faith abiding in the hearts of the 
faithful. All the syllables are of equal length, except the last. The two syllables 
tu-are so different in nature that they should not sound alike in rendition. The 
-tu, in spiritu, as the last syllable of a word, is soft; the tu-, of tuo, as the .first 
syllable of a new word and also as an accented syllable, is sung rather loudly 
and briskly. It is also the climax of the entire phrase. There is, then, a gradual 
crescendo to this syllable and a decrescendo from it. The last syllable, -o of tuo, 
is begun softly and is prolonged into silence. 

2 The Amen, too, should be rendered in a lively, brisk manner, the first 
syllable being the beginning of a word and an accented syllable, is brief and 
rather loud; the second syllable, as the end of a word is long and soft, tapering 
off into silence. Furthermore, since it is unlawful to accompany the priest's 
singing with the organ, it is inconsistent to accompany the choir or congregation 
when it gives the response. It is quite impractical, too, for the organist must 
often delay the reply until the right key is found. There should be no delay. 
The Et cum spiritu tuo should follow immediately upon the Dominus vobiscum 
not unlike grateful replies which follow heartfelt greetings in our social life. 
The Amen, too, should be intimately joined to the priest's prayer as a lusty 
"aye" follows a desired resolution. Our meetings in God's House should manifest 
every sentiment of joy and love. There is no choir or congregation so deficient 
in the musical art that it cannot sing the simple responses without the aid or 
distraction of the organ. 



reply, 'Amen', which means 'Let it be so as you ask' and 'That which 
you say is true.' In this simple word is embodied whatever the 
priest has said in many words and that word can be spoken with 
such devotion that he who speaks it obtains no less merit than if 
he had spoken all. For the Lord, our God, does not regard the mul- 
tiplicity of words so much as the fervor of our devotion." 1 

278. The School of the Mass. When the priest arrives at the 
altar, the choir welcomes him with the singing of the Introit. While 
this is a song of welcome to their spiritual director, it also sounds 
the keynote for the celebration of the day. It announces the reason 
for the gathering on this day. The Introit is composed of an antiphon 
followed by a verse of a psalm, to which is added the Gloria Patri. 
The antiphon is repeated. Meanwhile the priest is engaged in a 
private preparation for Mass at the foot of the altar. After the In- 
troit the choir (or better, the congregation) sings the Kyrie and 
Christe eleison. These Greek words can be translated: "Lord and 
Christ, have mercy on us." This nine-fold plea for pardon reminds 
us of God's infinite mercy and our infinite misery; it thus disposes 
us for the intimate union which we as sinful creatures are to enter 
with God, our sinless Creator. 

279. Before the "chairman" inaugurates the business of the 
day, however, he intones the Church's "Star-Spangled Banner," 
the Gloria in excelsis Deo. This corresponds to our national anthem 
as sung before all civic and social meetings. Because the priest is 
the chairman of the meeting, he may choose the tone of the hymn. 
Consequently, it is impolite for the choir to sing any other than that 
chosen by the priest. The priest, however, may not intone any but 
a Gregorian Gloria. 2 

280. After the Gloria the priest faces his people and greets them 
for the first time. Immediately after the response he states a propo- 
sition in the word Or emus which can be translated, "Let us ask God 
for something." The requests are always consistent with the feast. 
The people approve the priest's prayer in the word, Amen. 

281. Now follow the lessons of the day in the readings from the 
Old and the New Testament. In the older masses the so-called 
epistle was always a selection from the Old Testament. Today, 
especially on Sundays, selections are chosen from parts of the New 
Testament other than that of the four gospels. The Gradual follows 
this instruction. It is a joyful melody expressing thanks for the 
instruction received. 3 

l De Ponte, S. J., De christianis hominis perfectione (Coloniae: Agrippinae, 
1625). Quoted by Gihr, The Holy Sacrifice of the Mass (St. Louis, Mo. :Herder, 
1933) p. 425. 

2 There should be some understanding between the priest and the choir as to 
what Mass he intends to sing. The organist should give the priest the proper key 
by playing the intonation of the Gloria for him. Then the choir will be able to 
begin at the Et in terra pax without hesitation. While this is practical in that it 
saves time, it is also fitting in that it lends dignity to the celebration. 

3 It was sung in former times on an elevated platform; hence, the name Grad- 
ual, meaning 'a step'. 



282. The Alleluia and its versicle which follows the Gradual is 
a preparation for the reading of the New Testament selection. In 
former years a procession was held from the Epistle side of the altar, 
through the church and back to the Gospel side, during which the 
versicle was sung alternately with the verses of an appropriate 
psalm. The Gospel book was carried between two lighted candles 
at the head of the procession. We still have with us a remnant of 
this custom in the short procession from the Epistle to the Gospel 
side of the altar in the solemn Masses. On the Sundays of Lent and 
on feasts characterized by penance, a Tract takes the place of the 
Alleluia. During the Easter season, there is no Gradual. 

283. The portion of the Mass between the Epistle and the 
Gospel is indeed the musical part of the Mass. The most ornate 
melodies are sung; the priest performs no rite: he sits and listens. 
During most of the other selections the priest is attending to pre- 
scribed functions but during this part the Church asks him, as it 
were, to relax and to enjoy Her short musical program. 

284. The priest -greets his people before the Gospel as well as 
before the last Gospel of the Mass, although this is not a beginning 
of a new part. Perhaps the Church wishes to remind us that the 
New Testament is the era of God's living with us, the fulfillment of 
the prophecy: "I shall pitch my tent in your midst." (Lev. 26:11). 
Then again, the Dominus vobiscum may be sung to add solemnity 
to the New Testament reading or to prepare us for the reception of 
the good tidings. The priest announces the evangelist from whom 
the day's reading is taken in the words Sequentia sancti evangelii 
secundum . . ., meaning, "Continuation of the holy Gospel according 
to . . .", to which the people gratefully reply Gloria tibi, Domine, 
"Glory to Thee, Lord." At the conclusion the altar boy, but not 
the people or the choir, says, Laus tibi, Christe. Formerly, however, 
all said, Amen. After the Gospel the priest may give a homily, an 
explanation of the Gospel just sung, or a sermon on another topic. 

285. The priest then intones the Credo in Gregorian which the 
choir ought to continue in Gregorian chant. This is a mere profession 
of faith which ought not to be sung in a dramatic or theatrical 
fashion. It is customary in some churches for the priest to sing the 
Credo with the people. Such a custom is practical for three reasons: 
(1) The priest is vocally united to his people while the profession 
is made; (2) The Gregorian melodies are so plain that the difference 
between reciting them and singing them is small; (3) It dispenses 
with the rather clumsy rubric of having the people genuflect during 
the priest's recitation and kneel during the choir's singing of the 
words Et incarnatus est. There is no place for dramatics in a Gre- 
gorian Credo. In figured music where the composer delights in 
dramatizing the simple profession of faith, the basses will sing the 
words, "Was crucified, died and was buried" and then, with a 
sudden outburst the sopranos sing the joyful "On the third day He 
arose." The Popes have constantly condemned with justice features 
of the theater in church : they desire to preserve the House of God 


MASS AND THE CHAHT 1f 286-288 

as it should be kept: a place of holy peace and solemn prayer. The 
singing of the Credo concludes the school of the Mass. In ancient 
times the catechumens, those under instruction to become Christians, 
were dismissed. 

286. The Offertory Service. The priest faces his people and 
greets them again at the beginning of the offertory service. Then he 
sings the word Oremus, "Let us pray." The plural "us" is used even 
though he performs the act silently. Herein is clearly seen the official 
and public character of the priesthood. At this part of the Mass the 
priest offers to God the bread and wine which is later changed into 
the Body and Blood of Christ. He extracts them from all worldly 
use and makes them the property of God alone. During the Offer- 
tory the choir sings the proper Offertory of the Mass. Since the 
sixteenth century it has been greatly reduced in length. The only 
Offertory in its original grandeur is that of the Requiem Mass. In 
former years the Offertory was sung while the people went in pro- 
cession to the altar to offer their gifts for the sustenance of the 
priest and the poor. Today the ordinary collection, the people's 
contribution, is taken up at this time. But before the priest ends 
this service he asks the people's consent and approval by singing 
the last words of the last Secret prayer, Per omnia saecula saecu- 
lorum, to which the people reply with the customary Amen. 

287. The Thanksgiving Service. Again the priest greets the 
people with the familiar Dominus vobiscum. We are now entering 
the most sublime part of the Mass, for which the priest bids his 
people "Lift up your hearts" (Sursum corda) to which they reply, 
"We have lifted them to the Lord" (Habemus ad Dominum) . Then 
he irtvites the people to the service now to be performed, singing, 
"Let us give thanks to the Lord our God" (Gratias agamus Domino 
Deo nostro) to which the people consent and approve, answering, 
"It is worthy and just" (Dignum et justum est). Now follows the 
Preface in which the priest, speaking to God, proclaims that "it is 
truly worthy and just, right and helpful for salvation that we always 
and in all places give thanks to Thee, holy Lord, Father almighty, 
eternal God." If the Preface is proper for the day or season, the par- 
ticular reason is expressed why we should especially be thankful 
at this time. The Preface generally ends with the prayer that God 
allow us to join the choir of angels and heavenly citizens and with 
them say without end "Holy, holy, holy, Lord God of Hosts," etc. 
This is the Sanctus of the Mass. We should realize therefore that 
in singing the Sanctus we in this world have joined our voices with 
those of the Angels and Saints in heaven and with them seek to 
praise God in this heavenly song. 

288. There is a divergence of opinion among authorities as to 
whether the Sanctus should be sung in full before the Consecration. 
From the standpoint of opposing arguments, some recent declara- 
tions of the Holy See state that the Sanctus is to be sung to the first 
Hosanna in excelsis before the Consecration and the Benedictus 
after it. This appears, however, to be an answer to the question: 



Is it better to sing during the Consecration or to divide the Sanctus 
in two parts< observing silence at the Consecration? The words of 
a decree issued Nov. 12, 1831 read: 'The Benedictus should be 
sung after the Elevation of the Chalice." On May 22, 1894 the Holy 
See decreed : "No singing is allowed during the Elevation." Of course 
many of the modern Sanctus melodies are so long that they cannot 
be completed before the Elevation. No Gregorian Sanctus is so long 
that it cannot be completed before this sacred act. A further argu- 
ment against the singing of the Sanctus, in full, before the Conse- 
cration can be found in the Liber Usualis, under the title of "Ru- 
brics for the Chant of the Mass", No. VII. It states: "When the 
Preface is finished, the choir goes on with the Sanctus, etc., but ex- 
clusive of the Benedictus qui venit. Then only is the Elevation of 
the Blessed Sacrament. Meanwhile the choir is silent and adores 
with the rest. After the Elevation the choir sings the Benedictus." 

289. Arguments in favor of the complete Sanctus are given as 
follows: (1) The priest must recite the entire Sanctus without in- 
terruption. (2) The Liber Usualis in its melody gives no indication 
of an interruption between the first Hosanna and the Benedictus. 

(3) Pius X in his encyclical clearly states: "According to the laws 
of the Church, the Sanctus of the Mass must be finished before the 
Elevation, wherefore in this point the celebrant must attend to the 
singers." This last clause apparently means that the celebrant must 
proceed slowly in order to enable the choir to complete the Sanctus. 

(4) In former times it was expressly prescribed that the priest should 
sing the Sanctus with the people and that he should not begin the 
Canon until the Sanctus is completed. Each course seems to be 
supported by sufficient arguments. Whether the Sanctus is sung in 
full or in part before the Elevation, a motet in honor of the Blessed 
Sacrament may be sung in Latin after the Elevation provided it 
does not unduly delay the priest. 

290. The prayers of the thanksgiving service may be found in 
any missal. It does not lie within the scope of this book to explain 
them here. Even a cursory study of the prayers, however, will lend 
much to our devotion and appreciation at Mass. 

291. The thanksgiving service ends with the words said silently 
by the priest, words full of deep and holy meaning: "Through Him 
and with Him and in Him is to Thee, God the Father Almighty in 
unity with the Holy Ghost all honor and glory through all ages." 
These words signify that whatever we do with Christ and in Christ 
and through Christ during our lives gives honor and glory to God. 
To receive the people's approval of what has been done in his and 
their name, the priest sings the last of this prayer in an audible 
voice: Per omnia saecula saeculorum. The people reply Amen, agree- 
ing and consenting that it should be so. Thus ends the thanksgiving 

292. The Banquet. Without greeting the people but speaking 
directly to God in the name of the people, the priest and the people 



"are bold to say" the Our Father, or in Latin, the Pater noster. In 
this beautiful prayer clothed in plain but devout music the priest 
asks our Father in heaven for our daily bread. The people affix the 
ending plea Sed libera nos a malo, "but deliver us from evil", to 
which the priest silently adds Amen. 

293. Two requests are woven within this part of the Mass — 
requests that are intimately allied and complementary: pardon and 
peace. In the prayer immediately following- the Pater noster, the 
priest asks God to deliver us from all evils, past, present and future, 
and to grant us peace in our day. This prayer ends with the words 
Per omnia saecula saeculorum to which the people reply Amen. 
Then the priest recites "The peace of the Lord be always with us," 
Pax Domini sit semper vobiscum to which the people reply Et cum 
spiritu tuo. Again the congregation asks for pardon and peace in the 
Agnus Dei, "Lamb of God who takest away the sins of the world, 
have mercy on us" and "grant us peace." Sin divides us, grace 
unites us; where there is unity there must be peace. 

294. At the present time the Communion hymn is sung after 
the priest restores the Blessed Sacrament to the tabernacle. Formerly 
it was sung while the people received the Bread of Life. Its present 
form is nothing more than an antiphon which was originally sung 
alternately with appropriate psalms. It generally embodies the 
thought of the day's Mass incorporated into a short prayer of thanks- 
giving. The Amen at the end of the Postcommunion Prayer marks 
the end of the Communion Service. 

295. The Dismissal. In ordinary meetings the chairman an- 
nounces the completion of the session with the words: "The meeting 
is adjourned." In the liturgical parlance the announcement is made 
in the words Ite, missa est, "Go, the Mass is completed", to which 
the choir responds Deo gratias. In less joyful Masses where there is 
no Gloria in excelsis the words Benedicamus Domino, "Let us bless 
the Lord" or in Masses for the Dead, Requiescant in pace, "May they 
rest in peace" are substituted. But before the people disperse the 
priest gives his blessing and reads the beginning of St. John's gos- 
pel, a reading which was and still is attached to many of the Church's 
blessings. In it we are reminded that we are children of God through 
the Word that was made flesh and dwelt among us. The Church 
seems to imply that since we are children of God in Jesus Christ, 
we should go out to our daily duties and should live that life of 
whose fullness we have all received. 

296. It is often said that a Gregorian Mass is long. It may sur- 
prise many to learn that the difference between a low Mass and a 
Gregorian High Mass should not exceed ten minutes. If an ordinary 
low Mass is read in twenty-five minutes a High Mass need not ex- 
ceed thirty-five minutes. In substantiation of this the tabulation 
below shows relative differences in time between the Easter Mass 
and the ordinary Mass. The Easter Mass is longer than the usual 
Mass for it contains a Sequence. The singing of the melodies other 
than those noted should not detain the priest. 




Time required 
in singing 

Time required 
in reciting 

(delays priest) 


3 minutes 

3/4 minute 

2-1/4 minutes 









Total difference : 8-3/4 minutes 

In our busy day there seems to be no time for these few extra 
moments of worship, although we waste hundreds of such short 
moments throughout the day without feeling the least concern 
about our extravagance. A solemn Requiem High Mass sung as it 
should be with the whole of the Gradual, Tract and Dies irae should 
not extend over thirty-five minutes. Other Masses are no longer even 
with a Gregorian Gloria and Credo. Ornate or figured or non- 
Gregorian Gloria or Credo melodies consume the most time at a 
Mass. When one part is unduly prolonged we sometimes find it 
necessary to abbreviate another or omit it entirely. On the contrary, 
devotion should not be sacrificed to haste. If a Mass sung as pre- 
scribed should be extended five minutes, the spiritual gratification 
derived from the praise of Our Lord should more than compensate 
us for the loss of time. 


1f 297-299 


297. From the earliest ages of man, songs and melodies have been 
intimately associated with prayers and praises offered to God. 
Indeed, words alone do not suffice to express the desires of the heart : 
words are able to reveal the plain thoughts of the mind but when 
these thoughts are accompanied by deep feelings of the heart, they 
are expressed in music rather than in plain speech. Furthermore, 
when people gather together to offer a common service to God, or 
rather, to honor God as their common Father, their thoughts and 
sentiments must be expressed through a common and universal 
medium ; and that medium has ever been song, — song which dresses 
the words in garments worthy of the divine Majesty. It is only 
within the last few centuries that this garment has been laid aside. 
Then, while we laid aside the garment, instead of worshipping God 
as our common Father in plain words, we have ceased to worship 
Him at all. In consequence we have suffered the loss of the graces 
of love and fellowship which God gives when two or more are united 
in His name ; with the loss of grace comes the vices of hatred and war. 

298. After the Jews were firmly established and solidly united 
in the Promised Land, Solomon built a Temple wherein God's 
chosen people could offer a common sacrifice and could sing His 
praises. This sacrifice consisted of various religious acts accompanied 
by melodious prayers, for the most part, the singing of the psalms 
of David. Synagogues were built later in which the Jews gathered to 
learn the Law and to worship God in prayer and song. In addition 
to the sacrifices performed at the synagogues, sacrificial acts were 
performed also in the home where the father of the family was the 
presiding officer. On all occasions psalms were sung commemorating 
the rite and praising God for His goodness. 

299. Holy Scripture states that our Lord "went every year to 
Jerusalem at the solemn day of the pasch" and "performed all 
things according to the Law of the Lord" (Luke 2:41, 39). At these 
festivities He joined in the singing of the psalms with the others of 
His race in the praise of His heavenly Father. Even though Holy 
Scripture tells us but once that Jesus and His disciples "singing a 
psalm went forth to the Mount of Olives" (Mark 14: 16), neverthe- 
less, our Lord as a frequent visitor to the Synagogue and as a Jew 
faithful to the tradition of his people, "did join with His disciples 
in singing the traditional psalms." 1 

•Nielen, The Earliest Christianity (St. Louis, Mo. :1941). 



300. The beautiful canticle that sprang from the heart of the 
Blessed Mother at her visit to Elizabeth (known in the liturgy as 
the Magnificat), that Zachary sang at the birth of his son St. John 
the Baptist (known as the Benedictus) and that of Simeon as he held 
the Light of the world in his arms (known as the Nunc dimittis) are 
so like the psalms in structure, content and inspiration that we may 
rightly conclude that the singers of the New Testament were in- 
timately acquainted with them and held them dear to their hearts. 

301. The singing of the psalms accompanied all the religious 
acts of the Old Testament whether they were performed in the 
Temple, in the synagogue or in the home. The early Christians, 
whether converts from Judaism or paganism, carried this practice 
over to their own churches. The earliest Christians were converted 
Jews. We know that St. Paul in visiting new cities used the syna- 
gogues as meeting places to make contact with the Jews. He too 
participated in the old ritual of singing the psalms for, to him and 
his converts, Judaism was but the foreshadow of Christianity. They 
held firmly to the principle that Christ came not to destroy the law 
but to fulfill it. Consequently, although many of the Jewish cus- 
toms were abolished, as for instance the uncleanliness of meats and 
the observance of the Sabbath, nevertheless many were retained and 
clothed in the garb of the New Testament. In the Old Testament 
Christ was honored as the Messiah who is to come; in the New 
Testament He is honored as the Messiah who has come. For this 
reason many of the liturgical rites performed by the new converts 
under Jewish law are still maintained in our liturgy today. 

302. The Acts of the Apostles which tells the history of the first 
years of Christianity gives many an account of the meetings of 
Christians in which they "day by day persevered with one accord 
in the Temple, and breaking bread in private homes, took food with 
joyful simplicity of heart, praising God" (Eph. 5:19). St. Paul urges 
them to speak to themselves in psalms, and hymns and spiritual 
canticles, singing and making melody in their hearts to the Lord 
(cf. Col. 3:16). St. James advises: "Is anyone glad at heart? Let him 
sing a hymn." (5 : 13). St. Peter quoting a psalm says: "My heart is 
glad and my tongue rejoicest." (Acts 2 : 26). Thus it is clear that from 
the Temple of Solomon, through our Lord and His disciples, the 
custom of singing "psalms and hymns and spiritual canticles" has 
come down to us as a worthy, just, right and salutary means of 
giving praise and thanksgiving to God. 

303. The historical development of the chant parts of the Mass 
lies in the use of psalms in the early days of Christianity. The 
early Christians withdrew to the underground passages of the 
Catacombs and there spent many hours, at times whole nights, 
praising God in psalms and hymns in preparation and thanksgiving 
for the salutary gifts they received through the sacred mysteries. 
They sang the psalms antiphonally, that is, each part of an equally 
divided congregation sang an alternate verse. Later when they were 
permitted to celebrate the mysteries in public without the threat of 



persecution and molestation (313 A.D.), the melodies became more 
elaborate. A verse of a psalm that was especially suited to the day's 
festivity was extracted from the psalm and adorned with festive, 
elaborate melody. The congregation, however, could no longer sing 
this ornate selection; thus the order of singing changed from what 
is known as antiphonal to responsorial singing. In this the congre- 
gation sang the verses of the psalm while the chanters interposed 
the decorated refrain or response between the versicles. At this time, 
too, the Alleluia was introduced into the liturgy through the in- 
fluence of St. Jerome. This was used as a "refrain" or response before 
and after the versicle song following the Epistle during Eastertide. 
Its last syllable is fitted out with ajubilus, an ornate melody com- 
parable to the Tyrolese yodel which is nothing more than a melodi- 
ous outburst expressing the superabundant and overflowing joy of 
the heart. 

304. In time the refrain (now known as the Antiphon) which was 
sung by one or more chanters grew melodically richer. But while the 
melody was enriched, the text was shortened. The modern Introit, 
for instance, is nothing more than an antiphon with one verse of the 
original psalm, followed by the Gloria Patri (called the Doxology) 
and the repetition of the antiphon. The Gradual, too, has become so 
ornate that it consists of nothing but the refrain and a verse or ver- 
sicle of the original psalm. This too formerly had a responsorial 
character in that the Gradual was repeated after the versicle. The 
custom of abolishing the repetition, however, was soon sanctioned 
by the Church. Its repetition today is optional. The only Gradual 
which still retains most of its former fulness is that in the Mass for 
the first Sunday of Lent. It consists of a verse of Psalm 90 as the 
Gradual, a versicle from the same psalm and finally the Tract: the 
entire psalm except verses 8-10. This psalm which is quoted by 
Satan, as is told in the Gospel, pervades the whole Mass. The words 
of the Introit, the Gradual and Tract, the Offertory and Communion 
are extracted from it. The Offertories and Communions with the 
exception of those of the Requiem Mass have all lost their respon- 
sorial character. In Vespers and in fact in all the Hours of the Divine 
Office, each psalm is preceded and followed by a so-called antiphon. 
In order to understand the choice of a psalm versicle for a special 
feast, it is generally necessary to read the whole psalm for the one 
verse cannot give an adequate concept of its fitness to the feast. 

305. The Commons of the Mass (formerly called "Ordinarium 
Missae"), especially the Kyrie, Gloria and Sanctus, are of very early 
origin. They appear in the Liturgy of St. Gregory and must have 
been composed, therefore, before the year 600. The Agnus Dei, 
however, was added about a century later. The Credo appeared in 
the Mass about 800 but was later abolished; under Benedict VIII 
about the year 1020 it was reinstated for special feasts. These chants 
are comparatively simple in melody and construction because they 
have always been sung by the congregation. 1 

1 The dates given in the Liber refer to the earliest manuscript in which the 
melody is found (e.g., X s. means 10th century). 



306. It has been shown that in the development of chant the 
psalms were sung alternately or antiphonally by our Lord and His 
disciples in the synagogues of Palestine. This custom was carried 
over into Christianity by the Apostles and the early Christians. 
The persecutions of the infant Church, however, were not conducive 
to great development in music. But after the edict of Milan issued 
by Constantine the Great in 313, the Church was free to sacrifice 
and to worship God publicly. More elaborate melodies were com- 
posed at this time in the various monasteries in the East and later 
in the West. We must bear in mind that there was limited contact 
among monasteries at that time: a melody composed in one was 
never heard or learned outside its birthplace. Moreover, the method 
of notation was merely a crude memory aid indicating when the 
melody rose or fell but not indicating the distance of the rise or fall. 
Since the conventional signs could not indicate the rhythmic pe- 
culiarities adequately there was only one way of transferring a 
melody from one place to another : one person memorized the melody 
perfectly and then taught it to another who in turn communicated 
his learning to others. 

307. By the time the great music-loving monk was elevated to 
the papacy under the title of Gregory I (540 - 604), many monas- 
teries were already flourishing in Europe. The sons of St. Patrick in 
those days took to founding monasteries and writing music with the 
same natural facility with which they take to politics today. The 
famous monastery of St. Gall in Switzerland was founded in 613 on 
the site of the burial place of the Irish monk Cellach, whose name 
has been latinized to "Gallus." The beautiful Salve Sancte Introit of 
the Blessed Mother Masses, as also the hymn A solis ortus cardine, 
were said to have been written in the fifth century by an Irishman 
named "Sheil" and the latinized "Sedulius." Even as late as the 
ninth century we hear of a Notker who was educated with Tuotila by 
Iso and the Irishman Monegall at the abbey of St. Gall. It seems 
likely that Irish monks not only composed but through their travels 
also spread most of the chant as we have it today. It is true that in 
St. Gregory's time all but twenty-four pieces of Gregorian music had 
already been composed. Even these twenty-four might have been 
composed before his time, although it cannot be proved that they 
existed before the year 600 1 . Whether St. Gregory himself composed 
melodies is still a matter of dispute. All authorities agree that it is 
historically certain that he called together the chanters of the various 
monasteries and had each teach the others the proper melodies, that 
he compiled the first book of melodies, that he standardized the 
melodies for use in Rome whence they spread throughout Christen- 
dom and finally that he sent chanters to various parts of Europe to 
teach the melodies. Thus if the authorized and proper music of the 
Church is known as "Gregorian Chant", we may feel assured that 
no one other than Pope Gregory is more deserving of the distinction 
of the name. 

1 Cf. Catholic Encyclopedia under "Plain Chant." 



308. During the eighth century Charlemagne urged Adrian to 
send chanters into his empire to teach his subjects the sacred music 
as sung and taught in Rome. Adrian sent two monks, Petrus and 
Romanus, to found a school of music at Metz in Germany. Romanus 
became ill near the monastery of St. Gall and sought refuge and 
succor from the monks while Petrus continued to his destination. 
The monks persuaded Romanus to stay with them after his recovery. 
One of the most famous manuscripts on the chant dates from his 
time. It contains the melodies written in crude neumatic nuances 
which are of utmost importance and held in highest esteem by 
the monks of Solesmes. The symbolic letters, known as "Ro- 
manian Signs", depicted the rhythm so clearly that the Solesmes 
monks found them most helpful in rediscovering the true nature of 
the rhythm of the chant. 

309. The monastery of St. Gall became famous for another con- 
tribution to the chant. In the year 870 Blessed Notker Balbulus 
(the Stammerer) wrote what are now known as Sequences. He 
adopted syllables and words to the long jubilus in the last syllable 
of the Alleluia and arranged them in poetic form. In time a special 
Sequence was written for every Sunday and feast-day of the year, 
but Pope Pius V eliminated all but four. The fifth, the Stabat Mater 
of the Mass of the Seven Dolors, was added later. The Sequences are 
still sung at Mass after the Epistle in the place of the jubilus. At 
the same time another monk, Tuotilo (whose Gaelic name was 
"Tullach"), a class-mate of Blessed Notker, put words to the notes 
of the Kyrie chants. These are known as Tropes. Perhaps his pur- 
pose was merely practical in the sense that it was easier to designate 
and memorize notes that are distinguished by different syllables. 
It seems, however, that the Tropes were never intended or used for 
liturgical purposes. The Masses as printed in the Vatican Edition 
still retain the name given the Trope, in addition to their special 
number. Thus, for instance, the Common of the Mass during the 
Easter season is designated as : 

I — Tempore Paschali. 

(Lux et origo). 

The words Lux et origo are the words of the original Trope. The 
whole was sung: "Lord, Light and Origin of the world, have mercy 
on us." We still retain the opening words of Tuotilas' Tropes. 

310. One of the greatest incentives to the spread of the chant 
was invented by a companion to St. Bernard, a monk named Guido. 
He maintained strong views about the chant and thereby incurred 
the unpopularity of his brother-monks. He was sent or possibly went 
of his own accord under the duress of petty jealousy from the mon- 
astery of St. Maurus near Paris to Pomposa in Italy. The same fate 
of an ambitious musician met him there. He then went to a mon- 
astery near Arezzo and there apparently he found peace. He set 
himself to the task of placing the neumes on horizontal lines, one 
known and designated as the 'do-line', the other as the 'fa-line'. He 



later added another line between these two and still another below, 
all of which formed the staff of four lines as we have it today. The 
fifth line of our modern staff was not added until the seventeenth 
century. The Pope, John XIX, was elated at this invention for in it 
he saw the means of perpetuating and propagating the chant mel- 
odies without entrusting them to fickle memories. He summoned 
Guido to Rome to teach his discovery to others. Because of ill 
health Guido had soon to leave the city. The monks who had for- 
merly brought about his dismissal from their monastery now wel- 
comed his return. But Guido decided upon Arezzo and remained 
there until his death. It was he who gave the notes of the scale their 
sol-fa names. He noticed that in the hymn Ut queant laxis for the 
second Vespers for the feast of St. John the Baptist each division 
began with a successively higher note. The syllables corresponding 
to the notes were given to the seven notes of the scale in this way: 

UT queant laxis 

REsonare fibris 
MIra gestorum 

FAmuli tuorum, 
SOLve polluti 

LAbii reatum, 
Sancte Joannis. 

The French changed ut to do which was more sonorous to them ; ti 
is the European si formed from the first letters of the last words 
Sancte Joannes. To avoid confusion of the sol-fa si with the alpha- 
betic C we changed si to ti. Guido d'Arezzo had a strong belief that 
melodies should not exceed the range of their mode by more than 
one note above or below. He consequently reduced the range of 
sixty-three Graduals to his own style and taste. 

311. Among the slight changes which the chant sustained during 
this period was the change of the dominants in Modes III and 
VIII. In the course of the centuries do grew so powerful that it 
relegated ti to the servile position of a 'leading tone'. Consequently, 
in Modes III and VIII which should naturally have (and did form- 
erly have) ti as their dominant, the melody shifted adopting do. 
Hence, do is known as the 'modern' and ti as the 'ancient' dominant 
of these modes (see par. 179). 

312. During the ninth century polyphony was introduced by 
Hucbald, a Benedictine monk of St. Amand, in what he called the 
organum. He popularized the custom of adding another melody 
running parallel in fourths or fifths to the cantus firmus of the chant 
melody. So was born our present-day polyphony. Until that time 
Church music was strictly monophonic. In the following centuries, 



however, polyphony gained in popularity, at first gently eclipsing 
its parent and later, when it reached its peak in the sixteenth cen- 
tury, totally overshadowing it. From that time onward the chant 
sank rapidly into oblivion until the nineteenth century when the 
monks of Solesmes revived it on the point of its last breath, as it 
were. The reasons behind the discarding of the chant lay perhaps 
in the spirit which fostered the Renaissance. During the Dark Ages 
men were content to live together as children of God. Their parish 
church and their cathedrals which stand today as unsurpassed 
monuments of architecture are symphonies in stone. In everything 
they saw God the Creator and their common Father and they de- 
lighted in gathering together and singing praises to His Name. They 
lived as they prayed and, having prayed together in song, they 
lived together as brothers and sisters in Christ and children of God. 
But when the new style of music was introduced, music which the 
ordinary folk could no longer sing, they lost the consciousness of 
their common heritage and became individualized. 

313. The Great Awakening or Rebirth is hailed by modern 
Humanists as "a certain period, about fourteen centuries after 
Christ, to speak roughly, in which humanity awoke, as it were, 
from slumber, and began to live." It was "the emancipation of the 
reason for the modern world . . . for the mental condition of the 
Middle Ages was one of ignorant prostration before the idols of the 
Church." "During the Middle Ages man lived enveloped in a cowl. 
He had not seen the beauty of the world or had seen it only to cross 
himself, and turn aside and tell his beads and pray. Like St. Ber- 
nard travelling along the shores of Lake Leman, and noticing neither 
the azure of the waters nor the luxuriance of the vines, nor the ra- 
diance of the mountains with their fobe of sun and snow, but bend- 
ing a thought-burdened forehead over the neck of his mule — even 
like this monk, humanity had passed, a careful pilgrim, intent on 
the terrors of sin, death and judgment, along the highways of the 
world, and had not known that they are sight-worthy, or that life is 
a blessing. Beauty is a snare, pleasure a sin, the world a fleeting 
show, man fallen and lost, death the only certainty, judgment in- 
evitable, hell everlasting, heaven hard to win. Ignorance acceptable 
to God as a proof of faith, submission and abstinence and morti- 
fication are the only safe rules of life — these were the fixed ideas of 
the ascetic medieval church." Thus John Addington Symonds speaks 
in "The Rise of the Renaissance." Yet this St. Bernard of the simile 
is none other than the composer of the Jesu dulcis memoria a poem 
in which he expresses a joy unknown to us moderns. In fact, such 
is his estimate and appraisal of those who built the great cathedrals 
and monasteries of Europe one of which is the Westminster Abbey 
in which his king is proud to be crowned amid the splendor of me- 
dieval ritual. He says further: "Science was born, and the warfare 
between scientific positivism and religious metaphysics was de- 

314. Indeed the Renaissance was a re-awakening, man's re- 
awakening as an individual as the pagans conceived him. It was 


1 315-317 GREGORIAN CHA^T 

also a rebirth in so far as man who had heretofore lived and acted 
and prayed together with his fellowman as a child of the heavenly 
Father, awoke to find himself a selfish, independent and self-centered 
individual who lived for and in and by himself. Such a rebirth not 
only delivered the blow that eventually shattered Christianity into 
protesting religions but it even loosened the ties which bound the 
members of the true religion together. Individual composers began 
to write melodies in individual style, — melodies for the most part 
so complex and involved that the layman could no longer partake 
in the singing. The plain chant was gradually discarded because it 
was a plain and common music, because it gave no opportunities 
for individual performances and aggrandizements. Since men could 
no longer worship God together, they ceased to live together as 
children of one God. 

315. It is not our intention to condemn the Renaissance un- 
reservedly since it brought about much good. Our present purpose, 
however, is to disclose that one of its effects upon mankind was the 
destruction of the community spirit. During the Middle Ages so 
ingrained was the spirit of community that "the pride of authorship 
was unknown. It was plain that a doctrine belonged not to him who 
expounded it but to the Church as a whole. To write a book and so 
make known the truth to one's neighbor was in a sense to practice 
one of the works of mercy." 1 The same may be said of building 
churches, or carving statues, of painting pictures or of composing 
melodies. The artists are still unknown to us because they did not 
set their individual names to the work they wrought : they produced 
not for personal honor but for the glory of God and the good of the 

316. In regard to science, which was born during this age, more- 
over, Male states: "In the Middle Ages the idea of a thing which 
man framed for himself was always more real to him than the actual 
thing itself, and we see why these mystical centuries had no con- 
ception of what men now call science. The study of things for their 
own sake held no meaning for the thoughtful man. How could it be 
otherwise when the universe was conceived as an utterance of the 
Word of which every created thing was a single word? The task of 
the student of nature was to discern the eternal truth that God 
would have each thing to express, and to find in even' creature an 
adumbration of the great drama of the Fall and the Redemption. 
Even Roger Bacon, the most scientific spirit of the thirteenth cen- 
tury, after describing the seven coverings of the eye, concluded that 
by such means God had willed to express in our bodies an image of 
the seven Gifts of the Spirit." 

317. It is true that science is a blessing but it is true also that 
unless man adopts religion in proportion to the strength he derives 
from science, he will use his strength to his own destruction. If he 
grows individually and selfishly, shedding his intimate associations 

^Iale, Religious Art in France in the Thirteenth Centurv (London: Dent. 
1913) p. 34. 



with his fellow-man, selling his birthright as a child of God for the 
gains of this world, his follies will lead him to destruction. Science, 
the glory of the modern age, has grown so strong that it is possible 
to carry through the air bombs weighing many tons and with them 
to lay waste in a few minutes towns and cities, churches and cathe- 
drals, abbeys and monasteries, in which people have lived and 
prayed together in peace and love for centuries before the Great 

318. Various attempts have been made in the last centuries to 
restore the chant to its original purity and to restore it to its original 
place in the liturgy. All attempts failed except that made by the 
monks of Solesmes which God in His goodness brought about in the 
following manner. 

319. As a youth Prosper Gueranger, an inhabitant of Sable in 
France, took walks to the nearby town of Solesmes and meditated 
among the ruins of an ancient Benedictine monastery. His vivid 
imagination reviewed for him the life of work and prayer that was 
formerly carried on within the walls of the monastery. Prosper 
Gueranger was ordained in 1827 at the age of 22. In 1831 the ruins 
of the abbey were put to sale. The following year Pere Gueranger 
with the aid of private donations purchased it. In the next year five 
priests entered the restored priory and began the life according to 
the rule of St. Benedict. 

320. It was Dom Gueranger's purpose, perhaps even from child- 
hood, to restore the Benedictine life to his native France. He longed 
to work, pray and sing as the monks of old did before him. His 
greatest and most famous task, therefore, was to learn how the 
monks themselves in the Middle Ages sang the chant. He sent his 
monks to every monastery in Europe to get a photographic copy of 
every manuscript that they could find. The result is the famous 
"Paleographie" library in the monastery of Solesmes consisting of 
photographs of every page of 365 manuscripts. These were studied, 
compared and analyzed ; from the conclusions principles were formed 
on which the monks built their theory of rhythm. During the few 
years since the re-establishment of the monastery the French Gov- 
ernment expelled the monks no less than four times: in 1880, 1882, 
1883 and finally in 1903 when they were compelled to leave the 
country. They took refuge on the Isle of Wight near England and 
there continued their researches. They have since returned to the 
Abbey but are living in their own property as tenants to a pious 
landlord who had bought the property from the Government (to 
whom it did not belong) and had invited the monks to return. 

321. Two Solesmes monks must be mentioned in order to make 
our story complete: Dom Pothier and Dom Mocquereau. Dom Po- 
thier was choirmaster at Solesmes in 1904 when Pope Pius X, a 
year after his famous Motu Proprio, asked him to come to Rome to 
head a Commission for the Preparation of the Vatican Editions. 
While at Solesmes he laid the foundation for his successor as choir- 



master, Dom Mocquereau. He built upon the work left by Dom 
Pothier and after careful scientific research of over twenty years 
published .his "Le Nombre Musical Gregorien" which is and shall 
perhaps always remain the Summa on Gregorian Chant. 


During the Golden Age of the Jewish nation David composed 
psalms and Solomon, his son, built a fitting Temple wherein the 
people could gather and sing God's praises. Christ and His Apostles 
carried this custom over to Christianity where it prevailed until 
the Renaissance. At the official abolition of the sacrifice of the Pass- 
over of the Old Law and the official institution of the sacrifice of 
the Mass of the New Law Christ prayed: "That all may be one 
as Thou, Father, art in Me and I in Thee . . . that all may be per- 
fected into one, that the world may know that Thou has sent Me 
and that Thou hast loved them, even as Thou hast loved Me." 
(Joh. 17 :22, 23). But the world has become less one since it ceased to 
worship God as one. While the world prayed together, it lived to- 
gether; since it ceased to pray in common, it ceased also to live in 
common. It is reasonable, therefore, that the Church, the champion 
of peace and love, has so jealously kept its medium of expression, 
its sacred music, and that it rejoiced so jubilantly at its recent 


If 322-323 



322. This chapter presents the laws, regulations and intentions of 
the Church in regard to sacred music. The entire chapter will contain 
only quotations taken from the encyclicals and pastoral letters 
issued by the Holy Father to the whole Church or by Ordinaries of 
dioceses to their respective priests. It is true that only the ency- 
clicals issued by the Holy Father are binding upon the whole Church. 
The letters of the Holy Father and the Ordinaries, however, while 
obliging only those to whom they are addressed, nevertheless give 
us an insight into the intentions of the Church and often clarify and 
explain subject matter referred to in general terms in the ency- 

323. Each quotation is followed by a capital letter which identi- 
fies the document from which it was extracted. Those bearing the 
letters A and B are from the encyclicals of Pope Pius X and Pius XI 
respectively. They are binding upon the whole Church ; the others 
have an instructive value for us. The letters and their references 
are as follows: 

A: Motu Proprio of Pius X, issued November 22, 1903. 

B: Apostolic Constitution of Pius XI, issued December 20, 1928. 

C : Letter of Pius X to Cardinal Respighi as Vicar of Rome, De- 
cember 8, 1903. 

D: Regulations for Sacred Music issued in Rome by the Cardinal 
Vicar, February 2, 1912. 

E: Letter of Pius X to Archbishop Dubois, as Archbishop of 
Bourges, July 10, 1912. 

F : Pastoral letter of Cardinal Dubois, Archbishop of Paris, October 
9, 1921. 

G: Letter of Pius XI to Cardinal Dubois, as Archbishop of Paris, 
April 10, 1924. 

H: Letter of Pius XI to Cardinal Dubois, Archbishop of Paris, 
November 30, 1928. 



I — Need for Reform 

324. "Since the Church has received from Christ her founder the 
office of guarding the sanctity of divine worship, She is certainly 
bound to direct what concerns rites and ceremonies, formulas, 
prayers and singing in order to regulate better the august and 
perfect service of the Liturgy so-called since this is preeminently 
the sacred action." (B) 

325. "It is of great importance that whatever is done to enhance 
and adorn the Liturgy should be controlled by the laws and precepts 
of the Church, so that the arts may serve divine worship as most 
noble ministers. ***And this has been effected especially in sacred 
music; for wherever these regulations have been diligently carried 
out, there the ancient beauty of an exquisite art has begun to revive 
and a religious spirit to flourish and prosper, and there also the 
faithful, imbued more deeply with a sense of the Liturgy, have 
found the habit of partaking more zealously in the Eucharistic rite, 
in singing the psalms and in the public prayers." (B) 

326. "Nothing should be allowed in the sacred building that could 
disturb or lessen the beauty and devotion of the faithful, nothing 
that could be a reasonable motive for displeasure or scandal, noth- 
ing especially that could offend against the dignity and holiness of 
the sacred rites, and that would therefore, be unworthy of the 
house of prayer or of the majesty of God." (A) 

327. "There certainly is a constant tendency in sacred music to 
neglect the right principles of an art used in the service of the 
Liturgy, principles expressed very clearly in the laws of the Church, 
in the decrees of general and provincial councils, and in the repeated 
comments of the Sacred Congregations and of the Supreme Pontiffs 
Our Predecessors." (A) 

328. "We think it Our duty to lift up Our voice without delay in 
order to reprove and condemn everything in the music of divine 
worship that does not conform with the right principles so often 
expressed." (A) 

329. "Wherefore in order that no one in the future may bring for- 
ward as an excuse that he does not rightly know his duty, in order 
that all possible uncertainty concerning the laws already made may 
be removed We consider it advisable to sum up briefly the principles 
that govern the sacred music of liturgical service and to re-present 
the chief laws of the Church against faults in this matter. And, 
therefore, We publish this Our instruction motu proprio et certa 
scientia and We desire with all the authority of Our apostolic 
Office that it have the force of law as a canonical code concerning 
sacred music and We impose upon all by Our signature the duty 
of most exact obedience to it." (A) 

Oj&V "Sacred music, as an integral part of the liturgy, belongs to 
the general object of the liturgy, namely, the glory of God and the 
sanctification and edification of the faithful. It helps to enhance the' 



beauty and splendor of the ceremonies and since its chief duty is to 
clothe the liturgical text which is presented to the understanding 
of the faithful, with fitting melody, its object is to render that text 
more efficacious, so that the faithful may thereby be more aroused 
to devotion and better disposed to gather the fruits of grace which 
flow from the celebration of the sacred mysteries." (A) 

331. "At first the novelty will surprise some; very likely some 
choirmasters or directors will not be quite prepared for it, but little 
by little things will right themselves and everyone will find in the 
perfect correspondence of the music to liturgical laws and to the 
proper character of the chanting of psalms a beauty and a rightness 
which they have never felt before." (C) 

332. "By putting the matter off the difficulty would not become 
less, it would become greater: since the thing has to be done, let it be 
done at once and firmly." (C) 

333. "To achieve our purpose, positive, vigorous and enlightened 
action on the part of both secular and regular clergy is absolutely 
necessary." (D) 

II — Qualities of Church Music 

Q34J "Sacred music must eminently possess the qualities that be- 
long to liturgical rites especially holiness and beauty from which its 
every characteristic, universality, will follow spontaneously. It 
must be holy and therefore free from all that is secular, both in 
itself and in the method it is performed. It must be an art since in 
no other way can it have that effect on the mind of those who hear 
it which the Church intends of music in her liturgy. It must be 
universal in this sense that although each country may use whatever 
special forms may belong to its national style in its ecclesiastical 
music, these forms must be subject to the proper nature of sacred 
music as never to produce a bad impression on the mind of any 
stranger." (A) 

^35 o "To achieve its purpose, which is to enhance the solemnity 
of the offices and help to sanctify souls, this chant must be sacred, 
differing from profane tunes in inspiration, general character and 
method of execution. It must be grave like all that concerns divine 
worship, inducing recollection, closing the eyes, so to speak, to 
outward things and opening the heart to supernatural influences; 
it must be impressive, giving the soul a voice in which to utter its 
praise and prayer and adoration, and echoing that interior world 
which is in each one of us, and in which religious feeling vibrates so 
keenly at all times; it must be catholic, that is acceptable to all 
men of all races in all countries in every age, and finally it must be 
simple, with a simplicity which does not by any means exclude art, 
since a clear pure melody often expresses more beauty than the most 
learned and intricate musical compositions." (F) 



III— The Dignity of the Chant 

336. "These qualities (holiness, beauty and universality) are per- 
fectly found in Gregorian chant which is therefore the proper music 
of the Roman Church, the only music (a) which She inherited from 
the ancient Fathers; (b) which She has jealously kept for many 
centuries in her liturgical books ; (c) which She offers to the faithful 
as Her own music; (d) which She insists on being used exclusively in 
some parts of her liturgy, and (e) which, finally, has been so happily 
restored to its original perfection and purity by recent study." (A) 

337. "Plain chant has always been hailed as the highest model of 
Church music and We may with good reason set down as a general 
rule that the more a musical composition for use in church is like 
plain chant in its movement, its inspiration and its feeling, so much 
more is it right and liturgical, and the more it differs from this 
highest model so much the less is it worthy of the House of God." (A) 

338. "This ancient Gregorian chant should therefore be exten- 
sively restored in divine services and it should be understood that a 
service in church loses nothing of its solemnity when it is accom- 
panied by no other music than plain chant." (A) 

339. "Each part of the Mass and Office must keep even in its 
musical setting that form and character which it has from tradition 
and which is very well expressed in Gregorian chant." (A) 

340. "Is there indeed anything which exhales the fragrance of 
Christian beauty and fosters the Christian spirit like Gregorian 
chant?" (H) 

IV — Other Kinds of Music for Use in Church 
1. Classical Music. 

341. "The music of the classical school also possesses in a high de- 
gree the qualities described above, (holiness, beauty and universality) 
especially in that of the Roman School which reached its highest 
perfection in the sixteenth century under Pierluigi da Palestrina 
and which continued to produce excellent liturgical compositions. 
The music of this school agrees very well with the highest model of 
all sacred music, namely, plain chant, and it therefore deserves to 
be used in the more solemn offices of the Church as, for instance, in 
those of the Papal Chapel. It should also be largely restored es- 
pecially in the greater basilicas, in cathedrals, and in seminaries 
and in other institutions and where the necessary means of per- 
forming it are not wanting." (A) 

342. "We wish here also to recommend the formation of those ca- 
pellae musicorum or choirs which in course of time have come to be 
substituted for the ancient scholae and established in basilicas and 
larger churches especially to execute polyphonic music. In regard to 
this last point, sacred polyphony ought to be given a place only 
second to Gregorian chant itself, and on this account we earnestly 



desire that chgirs such as flourished from the fourteenth to the six- 
teenth century be renewed and revived today, especially in those 
places where the frequency and scope of divine worship demand a 
larger number of singers and more skill in the selection of them." (B) 

2. Modern Music. 

343. "More modern music may also be allowed in churches since 
good, serious and dignified compositions worthy of liturgical use 
have been produced. But since modern music has become a secular 
art, greater care must be taken when admitting it that nothing pro- 
fane is allowed, nothing that is mindful of the theatre, nothing that 
is based on the form of a secular composition." (A) 

344. "But we cannot refrain from lamenting that, just as for- 
merly, in the case of styles of music rightly prohibited by the Church, 
so today again there is danger lest a profane spirit should invade the 
House of God through new-fangled musical styles, which, should 
they get a real foothold the Church would be bound to condemn." 

V — Means of Reform 
1. General. 

345. "We desire all choirmasters, singers and clerics, all superiors 
of seminaries, ecclesiastical institutions and religious communities, 
all parish priests and rectors of churches, all canons of collegiate 
and cathedral churches and, most especially, the ordinaries of all 
dioceses, zealously to support these wise reforms which long were 
desired and unanimously hoped for in order that no injury be done 
to the authority of the Church which has already often proposed 
them and now insists on them once more." (A) 

346. "All higher schools of church music should be maintained in 
every way where they already exist and as far as possible new ones 
founded." (A) 

347. "It is most important that the Church should Herself pro- 
vide instruction for her own choirmasters, organists and singers, so 
that she may inspire them with the right principles of this sacred 
art." (A) 

348. "So that no one henceforth may seek easy excuses to con- 
sider himself exempt from the duty of obeying the laws of the Church, 
let all orders of canonical persons and religious communities discuss 
these matters at regular meetings; and just as formerly there was a 
cantor or director of the choir, so in future in choirs of canons and 
religious let some trained person be selected, not only to see that 
the laws of the liturgy and chant are put into practice, but also to 
correct the faults of individuals or of the whole choir. In this con- 
nection it must not be overlooked that according to the ancient and 
constant discipline of the Church, and in accordance with the cap- 



itular constitutions themselves, which are still in force, all who are 
bound to the choir office should be duly versed in Gregorian chant. 
And the chant to be used in all churches and all orders is that which, 
faithfully restored according to the old manuscripts has already been 
published by the Church in the authentic and standard Vatican 
Edition." (B) 

2. Bishops. 

349. "The bishop should appoint in each diocese a special com- 
mission of persons who are really competent in the matter, to whom 
they will entrust the duty of watching over the music performed in 
the churches in whatever way may seem most advisable. The com- 
mission will insist that the music is not only good in itself but also 
proportionate to the capacity of the singers so that they may always 
be well executed." (A) 

3. Rectors of Seminaries. 

350. "In ecclesiastical seminaries and institutions the traditional 
Gregorian chant must be studied with all diligence and love accord- 
ing to the law of the Council of Trent; and superiors should be 
generous in their appreciation and encouragement of this matter 
with their students." (A) 

351. "In the usual lectures on liturgy, moral theology and canon 
law, which are given to the students of theology, matters regarding 
principles and laws of sacred music must also be duly explained 
and means should be sought to complete this teaching with some 
special instruction on the aesthetics of sacred art so that the clerics 
may not leave the seminary without having right ideas on these 
subjects which are also a part of ecclesiastical knowledge." (A) 

352. "Those aspiring to the priesthood, not only in seminaries 
but also in religious houses, should be trained from their earliest 
years in Gregorian chant and sacred music because in childhood they 
learn more easily what belongs to melody, modulations and inter- 
vals, and any faults of voice can then be more readily eradicated, or 
at least corrected, whereas in later years they become irremediable." 

353. "In seminaries and other houses of studies there should 
therefore be for the due training of the clergy, brief but almost daily 
practice of Gregorian chant and sacred music. If this be carried out 
in the spirit of the liturgy, it will prove a solace rather than a burden 
to the minds of the pupils after the study of more exacting subjects." 

354. "But under no pretext whatsoever should less than two 
hours a week be devoted to the serious and practical study of sacred 
music, preferably the chant, and this in all institutions and for every 
pupil ; these two hours are not to include the time needed for choir 
practice." (C) 



355. "It is the express wish of His Holiness that in all institutions 
of ecclesiastical education — even those of regulars — great importance 
should be attached to the study of liturgical chant and sacred music 
as subjects of the greatest interest to the clergy." (C) 

356. "It is particularly important that Church students and 
young religious, should receive, in the course of their training in the 
Seminaries, Ecclesiastical Colleges and Religious Houses, a sound 
and thorough grounding in the liturgical chant and sacred music." 


4. Other Rectors. 

357. ' 'Those who are in charge of and take part in the public serv- 
ices in basilicas, cathedrals, collegiate churches and conventional re- 
ligious houses, shall make every endeavor to have the choir office 
duly restored and carried out according to the regulations of the 
Church; nor does this simply mean what is implied in the precept 
of reciting the office digne, attente et devote, but also what pertains to 
the art of singing." (B) 

5. Pastors. 

358. "Care should be taken to restore, at least in connection with 
the more important churches, the ancient choir schools which have 
already been introduced with very good results in many places. 
Indeed it would not be difficult for zealous priests to establish such 
schools even in small parishes and in the country, and they would 
form an easy means of gathering together children and adults to 
their profit and to the edification of all the parish." (A) 

359. "The superiors of churches and chapels, as also the Prefects 
of music in the chapters, must be well acquainted with the ecclesiasti- 
cal legislation relative to sacred music, and acquaint the choirmasters, 
organists and choristers with them, imposing and enforcing their 
obedience. It is they who, with the choirmaster, are to be held 
directly responsible for any transactions in the matter of music to 
be deplored in their churches." (D) 

360. "They are to see that the pieces selected are suitably ren- 
dered by a sufficient number of choristers capable of a performance 
worthy of art and liturgy; and this is why the singers ought to meet 
periodically for as many practices as may be deemed necessary. To 
insure this, however, both choirmasters and choristers must be 
properly paid. In the annual budget of each church a definite sum 
should be allocated to this purpose, and the expenses of showy 
feasts must be cut down so as to meet this end." (D) 

361. "In the courses of parochial instruction, or on other suitable 
occasions, they must expound the Holy Father's lofty purpose in 
reforming sacred music and invite the faithful to second their en- 
deavors chiefly by taking an active part in the sacred functions, 


f 362-367 GREGORIAN CHA^T 

singing in the common of the Mass (Kyrie, Gloria, etc.) as well as 
the psalms, the well-known liturgical hymns and the hymns in the 
vulgar tongue." (D) 

362. "To this end parish priests, rectors and superiors, especially 
in the larger churches, should apply all their zeal ; availing themselves 
of the help of some competent person to found their particular 
Schola cantor um" (D) 

363. ' 'The efforts of both secular and regular clergy, under leader- 
ship of their Bishops and Ordinaries, either working directly or through 
others especially trained for the task, should be devoted to the in- 
struction of their people in liturgical music, since this is so closely 
connected with Christian doctrine. This will best be accomplished 
by teaching Gregorian chant in the schools, pious sodalities and 
other liturgical associations. Moreover the communities of religious, 
whether men or women, should be eager to bring about this end in 
the educational institutions which have been entrusted to them." 

6. Choirmasters. 

364. "Since not only the rendering of the Gregorian chant, but 
also that of certain ancient and modern compositions is left entirely 
to the choir, there is a danger lest, both in the choice of pieces and the 
manner of singing them, the ecclesiastical regulations may be in- 
fringed. It is therefore necessary to make sure that all the members 
of the choir are technically competent and willing to observe each 
and every ecclesiastical regulation, and work for the application of 
the Pope's Motu proprio." (D) 

365. "Every Schola cantorum or choir should have its own spe- 
cial musical library for the ordinary performances in the Church, and 
they must possess first of all a sufficient number of Gregorian books 
in the Vatican Edition." (D) 

366. "Organists must take care in the accompaniment not to 
drown the voices by constant over-strong registration and by abuse 
of the reed-stops particularly; discretion is to be essentially observed 
in accompanying Gregorian chant. Even for the interludes and volun- 
taries they are to make use of approved written compositions." (D) 

7. Participation of the Faithful. 

367. "In order that the faithful may take a more active part in 
divine worship, let that portion of the chant which pertains to the 
Gregorian be restored to popular use. It is very necessary that the 
faithful taking part in sacred ceremonies should not do so as mere 
outsiders or mute spectators, but as worshippers thoroughly imbued 
with the beauty of the liturgy — and this even on occasions when pro- 
cessions and great functions are being held with clergy and sodalities 
present — so that they may sing alternately with the priest and the 
scholae, according to the prescribed rule. In this event we should 



not find the people making only a murmur or even no response at 
all to the public prayers of the liturgy, either in Latin or in the 
vernacular." (B) 

368. "It is a part of the authentic ecclesiastical tradition of the 
chant and sacred music that the entire assemblage of the faithful 
should take part in liturgical offices by means of this chant; following 
those portions of the text which are entrusted to the choir, while a 
special schola cantorum alternates with the congregation in render- 
ing the other parts of the text of the more complex melodies specially 
reserved for them." (D) 

369. "The grandeur of the sacred ceremonies increases in pro- 
portion to the numbers who join in them by singing." (G) 

370. "The faithful foregather at sacred shrines that they may 
draw piety thence, from its chief source, through actually partici- 
pating in the venerable mysteries and solemn public prayers of the 
Church." (B) 

8. Singers. 

371. "All liturgical singing other than that of the celebrant and 
sacred ministers at the altar belongs properly to a choir of clerics. 
Hence singers in church, if they are laymen, are substitutes for the 
ecclesiastical choir." (A) 

372. "Only men of known piety and integrity who, by their 
modest and reverent behavior during the sendee, show themselves 
worthy of the sacred duty they perform, may be allowed to sing in 
the choir." (A) 

373. "Scholae puerorum (junior choir schools for boys) should be 
encouraged not only in cathedrals and large churches but also in 
the smaller parish churches. The boys should be trained by the 
choirmaster so that, according to the old custom of the Church, they 
may join in singing in the choir with the men, especially when, as 
in polyphonic music, they are employed for the treble part which 
used to be called the cantus. (D) 

374. "Singers in church have a real liturgical office to perform and 
women being incapable of such an office, cannot be admitted to 
the choir. If high voices are needed, boys may be admitted according 
to the ages and custom of the church." (A) 

375. "Women may not sing in the liturgical functions unless 
amongst the people or representing them; they are therefore forbid- 
den to sing in tribunes or cantorias either alone, or more particularly 
as forming part of a choir. But religious living in community, and 
their pupils with them, may sing during the sacred functions in their 
own churches or oratories, in conformity with the decrees of the 
Sacred Congregation of Bishops and Regulars. At the same time 
they are absolutely forbidden to sing solos, and we desire that the 
preference should be given in singing Mass and Vespers, to Gregorian 
melodies rendered as far as possible by the whole community." (D) 



376. "Solos, while not entirely excluded, should never absorb the 
greater part of the liturgical text. They must rather be points of 
musical emphasis and accents bound up closely with the rest of the 
composition which should always remain strictly choral." (A) 

377. "The solo voice should never entirely dominate a sacred 
musical composition. It must only bear the character of a simple 
passage or melodic outline, strictly connected with the rest of the 
composition." (D) 

VI. Text 

378. "The language of the Roman Church is Latin. It is therefore 
forbidden to , sing anything in the vulgar tongue during solemn 
liturgical functions and much more is it forbidden to sing in the 
vulgar tongue the parts, either proper or common, of the Mass and 
of the Office." (A) 

379. "It is unlawful to change the words or their order, or to 
substitute another text or to omit entirely or in part except in the 
cases in which the rubrics allow the organ to replace certain verses 
which must then be recited in choir. Needless repetition is also 
forbidden." (A) 

380. "The perfection of Gregorian chant is closely bound up with 
the correct pronunciation of the words. Doubtless the melody is of 
itself independent of the text, yet they form but one thing in exe- 
cution. We may go further: the pronunciation of Latin words has 
exerted an active and often decisive influence on the formation of 
certain Gregorian phrases." (F) 

381. "The question of pronunciation of Latin is closely bound up 
with that of the restoration of Gregorian chant, the constant subject 
of Our thoughts and recommendations from the very beginning of 
Our pontificate. The accent and pronunciation of Latin has great 
influence on the melodic and rhythmic formation of the Gregorian 
phrase, and secondly it is important that these melodies should be 
rendered in the same manner in which they were artistically con- 
ceived at their first beginning." (E) 

382. " Tf the psalm prays, pray with it; if it weeps, weep; if it 
sings of joy, rejoice; if it speaks of hope, hope; if it expresses fear, 
fear.' (St. Agustine). How is-this possible, many will ask, since we do 
not know Latin? How is it possible, uttered in an unknown tongue? 
It is true that many do not understand the official language of the 
Church. Let those who can, at any rate, study and digest the sense 
of the words thoroughly. Their piety will profit thereby, and their 
singing gain greatly in beauty and expressiveness. As for the rest of 
the faithful, with a little good will they may arrive at the same 
results. The liturgical texts have been faithfully translated. Follow 
these translations and consult them. Read the English text of the 
Offices beforehand; you will soon become inbued with the general 
meaning of the words which you will have to sing." (F) 



383. "In Low Masses and Offices which are not strictly liturgical, 
such as triduums and novenas, and during the exposition of the 
Blessed Sacrament, singing is allowed even in the vulgar tongue, 
provided that both literary text and music have been approved by 
competent ecclesiastical authority." (D) 

VII. Rubrics 

384. "It is not lawful to make the Priest at the altar wait any 
longer than the ceremonies allow for the sake of the singing or 
instrumental music." (A) 

385. "It is only allowed to sing a motet in honor of the Blessed 
Sacrament after the Benedictus at High Mass. A short motet with 
words approved by the Church may also be added after the proper 
Offertory of the Mass has been sung." (A) 

386. "Let it be noted that it is not permissible to omit any one of 
the prescribed parts, common or proper, of the Mass, Office, or any 
other function. All the antiphons of the psalms and canticles, must 
be repeated all through when the rite requires it. When, as is some- 
times allowed, one portion of the liturgical text can be replaced by 
the organ, this text is to be recited in the choir in a voice which can 
be plainly heard and understood, or by the choristers themselves 
recto tono." (D) 

387. "During Low Masses solemnly celebrated, motets may be 
sung or the organ played, in accordance with the rubric. But this 
must be so contrived that the chants and organ playing are only heard 
when the priest is not reciting prayers aloud, that is, during the 
preparation and thanksgiving, from the Offertory to the Preface, 
from the Sanctus to the Pater Noster and from the Agnus Dei to 
the Post-Communion. Voice and organ must cease during the reci- 
tation of the Confiteor and the Ecce Agnus Dei if Communion be 
given." (D) 

388. "We have to point out that it is in error to suppose, as some 
have done, that in non-liturgical or extra-liturgical Offices one may 
perform musical compositions in free style which have already been 
condemned or pronounced unsuitable for the liturgical Offices. It is 
only fitting on the contrary to insist upon a dignified and serious 
style for all music rendered in God's House in the course of any 
sacred function whatsoever; for that of the solemn liturgy, other 
and special rules are laid down." (D) 

389. "The hymns of the Church must keep their traditional form. 
It is not lawful, for instance, to compose a Tantum Ergo in such a 
way that the first verse is a romance, an aria or an adagio and the 
Genitori an allegro." (A) 

390. "The Tantum Ergo and Genitori before the benediction of the 
Blessed Sacrament must be immediately followed by the Oremus 
and the Benediction, as in the actual course of these ceremonies it 
is not permissible to sing anything either in Latin or in the vulgar 
tongue." (D) 



391. "The Kyrie, Gloria, Credo, etc. must represent the unity of 
their text in the music." (A) 

392. ' 'According to the laws of the Church, the Sanctus of the 
Mass must be finished before the Elevation ; wherefore in this point 
the celebrant must attend to the singers." (A) 

393. "The Gloria and Credo, according to Gregorian tradition, 
should be comparatively short." (A) 

VIII. Organ 

394. "Although the proper music of the Church is only vocal, the 
accompaniment of an organ is allowed." (A) 

395. "There is one musical instrument, however, which properly 
and by tradition belongs to the Church, and that is the organ. On 
account of its grandeur and majesty it has always been considered 
worthy to mingle with liturgical rites, whether for accompanying the 
chant, or, when the choir is silent, for eliciting soft harmonies at 
times." (B) 

396. "Since vocal music must always be predominant, the organ 
and instruments may only sustain but never crush it." (A) 

397. "All organ playing must be performed not only according 
to the character of the instrument but also according to the rules of 
sacred music as described above." (A) 

398. "It is unlawful to introduce the singing with long preludes 
or to interrupt it with impromptu interludes." (A) 

399. "Let that organ music alone resound in our churches which 
expresses the majesty of the place and breathes the sanctity of the 
rites ; for in this way both the art of organ builders and that of the 
musicians who play the organ will be revived and render good service 
to the sacred liturgy." (B) 

400. "On the Ferias and Sundays of Advent and Lent, except 
Gaudete and Laetare Sundays, no instrument whatever must be 
played, even simply as an accompaniment to the voices. Yet a dis- 
creet accompaniment is allowed if solely to sustain the voices, and 
this only when Gregorian chant is sung and in case of real necessity, 
acknowledged by us. The use of any instrument whatsoever even 
merely as an accompaniment to the voices remains absolutely for- 
bidden in the Offices of the last three days of Holy Week." (D) 

401. "In some Masses of Requiem the organ or harmonium is 
allowed, but only to accompany the voices. At Low Masses of 
Requiem no instrument whatever must be played." 



IX. Other Instruments 

1. General. 

402. "In any special case, within the proper limits and with due 
care, other instruments may be allowed but never without special 
permission of the Bishop of the diocese." (B) 

403. "As we have learned that attempts are being made in differ- 
ent places to revive a kind of music which in no way befits the sacred 
Offices, particularly on account of its immoderate use of instru- 
ments, We hereby declare that chant combined with orchestra is by 
no means considered by the Church as a more nearly perfect form 
of music or more suited to sacred things. It is proper that the voice 
itself rather than musical instruments should be heard in the church- 
es; that is, the voice of the clergy, singers and congregation. It 
must not be thought that the Church is opposed to the advance of 
musical art in preferring the human voice to any instrument, but 
no instrument, however excellent and perfect, can surpass the human 
voice in expressing the feelings of the soul, most of all when it is 
used by the mind to offer prayer and praise to Almighty God." (B) 

2. Piano. 

404. "The use of the piano is forbidden in churches as also that 
of all noisy or irreverent instruments such as drums, kettledrums, 
cymbals, triangles, etc." (A) 

3. Bands. 

405. "Bands are strictly forbidden to play in Church and only for 
some special reason after the consent of the Bishop has been ob- 
tained may a certain number of specially chosen wind instruments 
be allowed; and the music they play must always be reverent and 
appropriate and in every way like the organ." (A) 

406. "Bands may be allowed by the Bishop in processions outside 
the church as long as they do not perform secular music." (A) 

407. "Without special permission, to be applied for on each sepa- 
rate occasion from the Apostolic Visita, no instrument except the 
organ or harmonium is to be played in the Church and notice is hereby 
given that it is not our intention to grant such permission except in 
altogether exceptional and peculiar circumstances. Authorization 
must also be sought each time that musical choirs wish to take part 
in outdoor processions and musical items must be confined to re- 
ligious pieces, expressly composed for the purpose, or better still to 
accompanying a hymn sung either in Latin or in the vulgar tongue 
by choristers or the faithful." (D) 



(All references are to paragraphs.) 

a, pronunciation of, 235 
Accelerando, 94 

nature in word, 96 

primary or tonic, 96 

secondary, 98, 108 

and ictus, 105, 109 

on rise note, 106 

on fall note, 107 

secondary melodic, 128 

brevity of, 143 

in modern music, 165 

in cadences, 203 

in chant, 381 

grammatical signs, 27 

and dynamics, 121 

in Romance languages, 96 

acute, 96 

grave, 96 
Accented beat, 163 
Accented syllable, 94-95 

in fall group, 133 
Accidentals, 8 
Adrian, Pope, 308 
ae, pronunciation of, 242 
Agnus Dei, 293 
agogics, 121, 126, 149, 152 
Alleluia, 303 

how sung, 270 

in Easter Vespers, 215 

in ordinary Vespers, 190 

at Mass, 282 
A men 

meaning of, 277, 290 

how sung, 270 
Anthems, 229 
Antiphon, 304 

at Introit, 278 

in Vespers, 191-192, 194, 196 
Arsis {see Rise Group) 

directions for singer, 17 

in Vespers, 191, 200, 206, 219 
au, pronunciation of, 242 
Authentic modes (see Modes) 

b, classification of, 253 
Bands in church, 405-407 
Banquet of Mass, 275, 292 ff. 
Bar line 

as divisions of the melody, 12 
and breathing, 12 
in modern music, 163 
Basic pulsation, 47 

Benedicamus Domino 

at Mass, 295 

at Vespers, 225 
Benedictus following Sanctus, 288-289 
Benedictine Monks of Solesmes 

restoration of the chant, 318 ff. 

and chironomy, 159 
Bishops, duties of, 349, 363 
Bivirga, 29, 32 

as climax, 134 
c, pronunciation of, 255 

nature of, 49 

spondees in, 207 

special dactylic, 212 

in Vespers, 198, 202, 207-212 

final, 198 
Capella musicorum, 342 
Capitulum, 217 
cc, pronunciation of, 262 
ch, pronunciation of, 262 

and the Mass, 271 ff. 

restoration of, 318 ff. 

dignity of, 336 ff. 

as art, 325 

regulations for, 324 ff 

place in liturgy, 330 

reform of, 324 ff. 

qualities of, 334-335 

lectures in, 351 

time for, 354 

conducting, 153 ff. 
Chapter, in Vespers, 217 
Charlemagne, 308 
Chironomy, 153 ff. 
Choirmaster, duties of, 345, 364 
Clef signs 

in general, 5 

do and fa, 6 

position of, 7 

history of, 310 
Clerics, duties of, 345 
climacus, 37 

rhythmic, 32 

in melody, 56 

in groups, 119-120 

in divisions, 124, 128, 149, 151-152 

in syllabic chants, 134(a) 

in psalms, 206 
clivis, 29, 31 
Comma, 13 


Commemorations, in Vespers, 223 
Communio (Communion hymn), 294 
Commission for Church Music, 349 

in Latin language, 232 

number of, 249 

classification of, 250 

production of, 254 

explosive, 251 

sustainable, 251 

voiced, 252 

unvoiced, 252 

place in the Mass, 285 

and text, 391 

length of, 393 
Crescendo, 94 
Custos {see Guide) 
d, classification of, 253 

definition, 97 

and natural rhythm, 103 

and accent, 104 

and rise to accent, 129 
Dactylic cadence {see Cadence), 103 
Decrescendo, 94 
Diphthongs, 240-241 
Directing the chant {see Chironomy), 

153 ff. 
Dismissal in Mass, 275, 295 
Division of melodies, 12-14 

nature of, 124 

how rendered, 152 
do clef, 6 
do line, 310 

Dominant of Modes {see Modes), 174, 
176, 182 

modern and ancient, 179, 311 

in psalms, 197 

in modern music, 172 

Dominus vobiscum, 216, 284, 287 

Dotted note, 49 

Double phrase mark (also see Bar line), 


of notes, 44 

in rhythm, 57, 60 

importance of, 60-61, 93 

of syllables, 94 

of pause in psalmody, 206 

definition, 121 

and rhythm, 126 

and expression, 149 
e, pronunciation of, 238 
ei, pronunciation of, 244 

appearance in print, 15 

function, 50, 145 
Epistle, 281 

eu, pronunciation of, 243 
E u o u a e , meaning of, 216 
excelsis, pronunciation of, 268 

Expression and dynamics, 149 

/, classification of, 250 

fa clef, 6 

fa line, 310 

Fall, in rhythm, 57 

Fall note (or ictus) 

qualities of, 62-63, 68 

and repose, 68 

kinds of, 74 

in division, 123, 144 

in melody, 57-58 

in stress, 57, 59, 164 

in duration, 57, 60, 167 

in elementary rhythm, 65 

in simple rhythm, 75 

in time groups, 75, 79-80, 82 

in groups, 85, 86 

on accented syllable, 102, 107 

on last syllable, 99, 103-104 

in neumes, 147 

as ictus in length, 148 

in modern music, 162 ff. 
Fall group (or thesis) 

nature of, 116-117 

qualities of, 118 

place for, 119 

in division, 123, 139, 145, 150 

how delineated, 154, 156 

Fall divisions 

in subordinate divisions, 124 
in syllabic chants, 134 

Final cadence {see Cadence), 198 
Final in modes {see Modes), 174, 176 

in psalmody, 198 
Flat {see te) 

flex, in psalms, 195, 214 
Foreign words, 269 
Full bar {see Bar line) 
g, classification of, 240, 256 
Glide-off vowels, 240 
Glide-on vowels, 240 
Gloria, 279 

place in the Mass, 279 

and text, 391 

length of, 393 
Gloria Patri 

in the Introit, 278 

in Vespers, 189-190, 193 
gn, pronunciation of, 264 
Gospel, 281, 284 
Gradual, 282-283 
Grave accent, 96 

Greek words, pronunciation of, 269 
Gregorian Chant {see Chant) 
Gregorian Mass, 296 
Gregory the Great, 307, 321 
Group of notes 

definition of, 75 

nature of notes in, 79 ff. 

nature of, 116 ff. 

retain characteristics, 123 

in syllabic chants, 134(b) 

in modern music, 169 


Grouping of notes 

in neumatic chants, 84-85 

in syllabic chants, 91 ff., 115 

examples of, 126-127, 129-130, 148 
Guide note, 11(f) 
Guido d'Arezzo, 310 
Gueranger, 319-320 
h, classification of, 257 
Half bar (see Bar line) 
Half tone, 3 

Hebrew words, pronunciation of, 269 
Hucbald, 312 

in general, 389 

in Vespers, 218 

symmetrical division in, 14 

Ut queant laxis, 310. 
i, pronunciation of, 239 

before u, 248 
ij, iij in Kyrie, 18 
Ictus {see also Fall note) 

definition, 72 

other terms, 73 

place of, 85 

cannot succeed each other, 85 

in word, 99, 104 

in syllabic chants, 99, 110 

on weak syllables, 155 

independent of accent, 105 

on accent syllables, 103, 107 

on final syllables, 99, 106, 164 

on unaccented syllables, 155 

in modern music, 162 

as sign of division, 12(d) 
unequal length of, 14 
beginning with isolated note, 89 
nature of, 12(d), 127, 130 
in practice, 125 ff. 

Instruments in church, 402-403 


in Vespers, 196 
of Magnificat, 220 

Introit, 278 
Isolated note 

how treated, 87 

at beginning of melody of phrase, 
88, 138 
lie missa est, 295 
j, pronunciation of, 245, 258 
John XIX, Pope, 310 
Joining of divisions, 137 ff. 
Jubilus, 282 
Keys for Gregorian melodies 

in relation to modern scale, 4 

choice of key, 185-186 

for psalms, 231 
Kyrie, 278, 391 
/, classification of, 253 
Latin pronunciation 

vowels, 234 ff. 

consonants, 249 ff. 

diphthongs, 240 ff. 

and foreign words, 269 

use of, 380-381 
Leading tone, 179 
Leaps in chant, 147 
Length of Gregorian note, 46 

signs of, 48 ff. 
Leger lines, 7 

Legislation of church music, 322 ff. 
Lessons at Mass, 281 
Liber Usualis, 189 
Life through rhythm, 77 
Liquescent notes 

appearance, 11(e) 

purpose of, 43 

definition of, 271 

nature of, 324-325, 330 
Low Masses, 383, 387 
m, classification of, 253 
Magnificat, 220 

Mass, liturgical division of, 275 ff. 
Measure in modern music, 82 
Mediant cadence, 198 
Melismatic chants, 19(c) 
Melodic climax, 119, 124, 130 

divisions of, 12, 14 

kinds of, 19 

in chant, 20-21 

natural melody of word, 111-112, 130 

Member, 12(c), 89 

nature of, 128, 131 
Men in choir, 371-372 
Mocquereau, 321 

in general, 172 ff. 

formula of, 173 

division of, 175-176 

transposed, 178 

study of, 182 

characteristics of, 180-181 

melodies of, 184 

Monosyllabic words, 204 
Mora vocis, 127 
Motets, 387 


sacred, 325, 327 ff., 33 
classical, 341 
modern, 343 

n, classification of, 253 
Natural sign, 10 
Neumatic chants, 19(b) 

in general, 24-27 

kinds of, 28-29 

of two notes, 29 

of three notes, 33 

compound, 42 

special, 43 

in selection, 143 
Non-Latin words, pronunciation of, 

Non-liturgical services, 388 



shape of, 1, 11, 62 

nature of, 123 

on staff, 2 

square, 11(a) 

inclined, 11(b) 

virga, 11(c) 

quilisma, 11(d) 

liquescent, 11(e) 

guide, 11(f) 

dotted, 49 

isolated, 87-88 

in modern music, 166 
Notker, Blessed, 309 
Novenas, 383 
o, pronunciation of, 236 
oe, pronunciation of, 242 
Offertory, 275, 286 
Orchestra in church, 403 
Ordinaries, duties of, 345, 363 
Oremus, 280, 286 
oriscus, 41 

use of, 394 ff. 

in Requiem Masses, 401 

at other services, 400 
Organist, duties of, 366 
p, classification of, 253 
Participation of the faithful, 272, 313, 

Palestrina, 341 

Pastors, duties of, 358, 360-361 
Pauses (see Duration, Mora vocis, 

Bar line) 
Peace in chant, 151 
Petrus, 308 

ph, pronunciation of, 257 

division sign, 12(b) 

and isolated note, 88 

nature of, 132 

completeness of, 141 

in modern music 168, 170 
Piano in church, 404 

of Gregorian notes, 4 

and rhythm, 57-58, 93 

choice of, 185 ff. 

Plagal modes {see Modes), 175-176 

Plain chant {see Chant) 

Plain song {see. Chant) 

podatus, 29-30 

Polyphonic music, 312, 342, 373 

porrectus, 39 

Pothier, 321 

Preface at Mass, 287 

Preparatory syllables, 203 


as sign of length, 52, 86 

in grouping, 130, 147 
Priest at Mass, 274 
Pronunciation of Latin, 232 ff., 381 
Psalmody, 188 ff. 

Psalms, structure of, 195 
Pulsation, basic, 47 
punctum {see Notes) 
qu, pronunciation of, 259 
Quarter bar (see Bar line) 

shape, 11(d) 

characteristic of, 35 

rhythm of, 53 
r, pronunciation of, 260 
rallentando, 94 
Range of modes, 174 
Reciting tone of psalms, 196-197 
Rectors of seminaries, duties of, 350, 

Renaissance, 312 ff. 
Relation of melody, text and rhythm, 

20, 45 
Religious communities, 375 
Repetition, as sign of length, 51 

in chant, 20, 23 

nature of, 55 ff. 

elementary, 65 

simple, 75, 126 

purpose of, 76 ff. 

and text, 100 

natural rhythm of word, 100, 113, 
129, 130 

in modern music, 161 ff., 171 


in rhythm, 57 
expresses activity, 67 
rendition of, 72 

Rise note 

nature of, 57 ff., 64 ff. 

qualities of, 62-63, 67 

rendition of, 72 

in pitch, 58 

in duration, 60 

in division, 123 

brevity of, 143-144, 146, 167 

direction of, 155-156 

in elementary rhythm, 65-66 

in simple rhythm, 75 

in time groups, 75, 79, 80, 82 

as isolated note, 87-88 

on accented syllable, 102, 104 

in neumes, 147 

in modern music, 162 ff., 167 
Rise group (or Arsis) 

essential elements, 116-117 

nature of, 118 

place for, 119 

in division, 123, 139-140, 145, 150 

how delineated, 154 

Rise divisions 

in neumatic chant, 124 

in syllabic chant, 134 
Romanus, 308 
Rubrics, 384 ff. 
s, classification of, 253 
St. Gall, 308-309 


Sacred music 

character, 330 

function, 325 

need for reform, 327 

qualities, 334-335 

decrees, 322 ff. 

character of, 36 

prolongation of, 53, 86 

for the leap, 146 
Sanctus at Mass, 287 ff. 
sc, pronunciation of, 265 

modern, 172 

Gregorian (see Modes) 

sol-fa, 3 

numbers, 2 
scandicus, 34 
sch, pronunciation of, 266 
Schola cantorum, 362, 365 
Schola puerorum, 373 
School of the Mass, 275, 278 
Sequence, 309 
Sharps in chant, 8 
Signs of division (see Bar line) 
Singers in church, 345, 371 
Solesmes, monastery of, 308-309 
Sol-fa notes, 2 

pitch of, 4 

scale, 3 

names, 310 
Solos in church, 376-377 
Spirituality in chant, 150 
Spondee, 97, 102, 104 
Staff, 1 

a feature of rhythmic movement, 57 

quality of, 59 

in words, 93, 107 
Suffrage in Vespers, 223 
Syllabic chants, 19(a) 

grouping of, 134 
Syncopation, 61 
t, classification of, 250 
Tantum ergo, 390 
te, 8 

influence of, 9 

indicated by, 10 

in modes, 173, 179 
Tempo (see also Agogics), 152 
Text in chant 

an essential element, 20 

reading, 22 

Latin, 378 

translation of, 382 

correct pronunciation of, 380 

recitation of, 386 
th, pronunciation of, 257 
Thanksgiving Service at Mass, 275, 287 
Thesis (see Fall group) 
ti (see te) 

as dominant, 311 
-ti-, pronunciation of, 267 
Time groups (see Groups) 
Tonality, 172 

Tonic in modern music, 172 
Tonus Peregrinus, 213 
tor cuius, 38 
Triduum, 383 
tristropha, 40 
Tropes, 309 
Tuotila, 309 
u, pronunciation of, 237 
Unaccented beat in modern music, 163 
Undulation, 133-134, 155 
Uniting notes 

in neumatic chants, 79 ff. 

in syllabic chants, 91 ff. 

groups, 115 ff. 

divisions, 137 
Unity through rhythm, 77 
Unvoiced consonants, 252 
v, classification of, 253 
Vatican Edition, 321 
Verses in psalmody, 206 
Versicles, 189, 218-219, 223, 229 

outline of, 189 

for Eastertide, 215 

for Advent, 216 
Vocal exercises, 185 (footnote) 
Vocal music, 394, 396, 403 
Voiced consonants, 252 

nature of, 232 

number in Latin, 233 

formation of, 234 

purity of, 248 

glide-on, 240 

glide-off, 240 
Women in choir, 374-375 
x, pronunciation of, 268 
y, pronunciation of, 246 
z, pronunciation of, 261